SU\G(𝔸)/K·U
Sign Up for free Log In

Author Description

Login to generate an author description

Ask a Question About This Mathematician

GroupIM

2020-07-25
Aravind Sankar , Yanhong Wu , Yuhang Wu +3 more
We study the problem of making item recommendations to ephemeral groups, which comprise users with limited or no historical activities together. Existing studies target persistent groups with substantial activity history, … We study the problem of making item recommendations to ephemeral groups, which comprise users with limited or no historical activities together. Existing studies target persistent groups with substantial activity history, while ephemeral groups lack historical interactions. To overcome group interaction sparsity, we propose data-driven regularization strategies to exploit both the preference covariance amongst users who are in the same group, as well as the contextual relevance of users' individual preferences to each group. We make two contributions. First, we present a recommender architecture-agnostic framework GroupIM that can integrate arbitrary neural preference encoders and aggregators for ephemeral group recommendation. Second, we regularize the user-group latent space to overcome group interaction sparsity by: maximizing mutual information between representations of groups and group members; and dynamically prioritizing the preferences of highly informative members through contextual preference weighting. Our experimental results on several real-world datasets indicate significant performance improvements (31-62% relative NDCG@20) over state-of-the-art group recommendation techniques.
View PDF Chat

"Birds of a Feather": Does User Homophily Impact Information Diffusion in Social Media?

2010-01-01
Munmun De Choudhury , Hari Sundaram , Ajita John +2 more
This article investigates the impact of user homophily on the social process of information diffusion in online social media. Over several decades, social scientists have been interested in the idea … This article investigates the impact of user homophily on the social process of information diffusion in online social media. Over several decades, social scientists have been interested in the idea that similarity breeds connection: precisely known as "homophily". Homophily has been extensively studied in the social sciences and refers to the idea that users in a social system tend to bond more with ones who are similar to them than to ones who are dissimilar. The key observation is that homophily structures the ego-networks of individuals and impacts their communication behavior. It is therefore likely to effect the mechanisms in which information propagates among them. To this effect, we investigate the interplay between homophily along diverse user attributes and the information diffusion process on social media. In our approach, we first extract diffusion characteristics---corresponding to the baseline social graph as well as graphs filtered on different user attributes (e.g. location, activity). Second, we propose a Dynamic Bayesian Network based framework to predict diffusion characteristics at a future time. Third, the impact of attribute homophily is quantified by the ability of the predicted characteristics in explaining actual diffusion, and external variables, including trends in search and news. Experimental results on a large Twitter dataset demonstrate that choice of the homophilous attribute can impact the prediction of information diffusion, given a specific metric and a topic. In most cases, attribute homophily is able to explain the actual diffusion and external trends by ~15-25% over cases when homophily is not considered.

Social event detection with interaction graph modeling

2012-10-29
Yanxiang Wang , Hari Sundaram , Lexing Xie
This paper focuses on detecting social, physical-world events from photos posted on social media sites. The problem is important: cheap media capture devices have significantly increased the number of photos … This paper focuses on detecting social, physical-world events from photos posted on social media sites. The problem is important: cheap media capture devices have significantly increased the number of photos shared on these sites. The main contribution of this paper is to incorporate online social interaction features in the detection of physical events. We believe that online social interaction reflect important signals among the participants on the "social affinity" of two photos, thereby helping event detection. We compute social affinity via a random-walk on a social interaction graph to determine similarity between two photos on the graph. We train a support vector machine classifier to combine the social affinity between photos and photo-centric metadata including time, location, tags and description. Incremental clustering is then used to group photos to event clusters. We have very good results on two large scale real-world datasets: Upcoming and MediaEval. We show an improvement between 0.06-0.10 in F1 on these datasets.
View PDF Chat

Transfer Learning via Contextual Invariants for One-to-Many Cross-Domain Recommendation

2020-07-25
Adit Krishnan , Mahashweta Das , Mangesh Bendre +2 more
The rapid proliferation of new users and items on the social web has aggravated the gray-sheep user/long-tail item challenge in recommender systems. Historically, cross-domain co-clustering methods have successfully leveraged shared … The rapid proliferation of new users and items on the social web has aggravated the gray-sheep user/long-tail item challenge in recommender systems. Historically, cross-domain co-clustering methods have successfully leveraged shared users and items across dense and sparse domains to improve inference quality. However, they rely on shared rating data and cannot scale to multiple sparse target domains (i.e., the one-to-many transfer setting). This, combined with the increasing adoption of neural recommender architectures, motivates us to develop scalable neural layer-transfer approaches for cross-domain learning. Our key intuition is to guide neural collaborative filtering with domain-invariant components shared across the dense and sparse domains, improving the user and item representations learned in the sparse domains. We leverage contextual invariances across domains to develop these shared modules, and demonstrate that with user-item interaction context, we can learn-to-learn informative representation spaces even with sparse interaction data. We show the effectiveness and scalability of our approach on two public datasets and a massive transaction dataset from Visa, a global payments technology company (19% Item Recall, 3x faster vs. training separate models for each domain). Our approach is applicable to both implicit and explicit feedback settings.
View PDF Chat

Inform the Uninformed: Improving Online Informed Consent Reading with an AI-Powered Chatbot

2023-04-19
Ziang Xiao , Tiffany Wenting Li , Karrie Karahalios +1 more
Informed consent is a core cornerstone of ethics in human subject research. Through the informed consent process, participants learn about the study procedure, benefits, risks, and more to make an … Informed consent is a core cornerstone of ethics in human subject research. Through the informed consent process, participants learn about the study procedure, benefits, risks, and more to make an informed decision. However, recent studies showed that current practices might lead to uninformed decisions and expose participants to unknown risks, especially in online studies. Without the researcher's presence and guidance, online participants must read a lengthy form on their own with no answers to their questions. In this paper, we examined the role of an AI-powered chatbot in improving informed consent online. By comparing the chatbot with form-based interaction, we found the chatbot improved consent form reading, promoted participants' feelings of agency, and closed the power gap between the participant and the researcher. Our exploratory analysis further revealed the altered power dynamic might eventually benefit study response quality. We discussed design implications for creating AI-powered chatbots to offer effective informed consent in broader settings.
View PDF Chat

Beyond Localized Graph Neural Networks: An Attributed Motif Regularization Framework

2020-11-01
Aravind Sankar , Junting Wang , Adit Krishnan +1 more
We present InfoMotif, a new semi-supervised, motif-regularized, learning framework over graphs. We overcome two key limitations of message passing in popular graph neural networks (GNNs): localization (a k-layer GNN cannot … We present InfoMotif, a new semi-supervised, motif-regularized, learning framework over graphs. We overcome two key limitations of message passing in popular graph neural networks (GNNs): localization (a k-layer GNN cannot utilize features outside the k-hop neighborhood of the labeled training nodes) and over-smoothed (structurally indistinguishable) representations. We propose the concept of attributed structural roles of nodes based on their occurrence in different network motifs, independent of network proximity. Two nodes share attributed structural roles if they participate in topologically similar motif instances over co-varying sets of attributes. Further, InfoMotif achieves architecture independence by regularizing the node representations of arbitrary GNNs via mutual information maximization. Our training curriculum dynamically prioritizes multiple motifs in the learning process without relying on distributional assumptions in the underlying graph or the learning task. We integrate three state-of-the-art GNNs in our framework, to show significant gains (3-10% accuracy) across six diverse, real-world datasets. We see stronger gains for nodes with sparse training labels and diverse attributes in local neighborhood structures.
View PDF Chat

SCENT

2011-10-01
Yu‐Ru Lin , K. Selçcuk Candan , Hari Sundaram +1 more
We propose SCENT, an innovative, scalable spectral analysis framework for internet scale monitoring of multirelational social media data, encoded in the form of tensor streams. In particular, a significant challenge … We propose SCENT, an innovative, scalable spectral analysis framework for internet scale monitoring of multirelational social media data, encoded in the form of tensor streams. In particular, a significant challenge is to detect key changes in the social media data, which could reflect important events in the real world, sufficiently quickly. Social media data have three challenging characteristics. First, data sizes are enormous; recent technological advances allow hundreds of millions of users to create and share content within online social networks. Second, social data are often multifaceted (i.e., have many dimensions of potential interest, from the textual content to user metadata). Finally, the data is dynamic; structural changes can occur at multiple time scales and be localized to a subset of users. Consequently, a framework for extracting useful information from social media data needs to scale with data volume, and also with the number and diversity of the facets of the data. In SCENT, we focus on the computational cost of structural change detection in tensor streams. We extend compressed sensing (CS) to tensor data. We show that, through the use of randomized tensor ensembles, SCENT is able to encode the observed tensor streams in the form of compact descriptors. We show that the descriptors allow very fast detection of significant spectral changes in the tensor stream, which also reduce data collection, storage, and processing costs. Experiments over synthetic and real data show that SCENT is faster (17.7x--159x for change detection) and more accurate (above 0.9 F-score) than baseline methods.

Discovering Strategic Behaviors for Collaborative Content-Production in Social Networks

2020-04-20
Yuxin Xiao , Adit Krishnan , Hari Sundaram
Some social networks provide explicit mechanisms to allocate social rewards such as reputation based on users' actions, while the mechanism is more opaque in other networks. Nonetheless, there are always … Some social networks provide explicit mechanisms to allocate social rewards such as reputation based on users' actions, while the mechanism is more opaque in other networks. Nonetheless, there are always individuals who obtain greater rewards and reputation than their peers. An intuitive yet important question to ask is whether these successful users employ strategic behaviors to become influential. It might appear that the influencers "have gamed the system." However, it remains difficult to conclude the rationality of their actions due to factors like the combinatorial strategy space, inability to determine payoffs, and resource limitations faced by individuals. The challenging nature of this question has drawn attention from both the theory and data mining communities. Therefore, in this paper, we are motivated to investigate if resource-limited individuals discover strategic behaviors associated with high payoffs when producing collaborative/interactive content in social networks. We propose a novel framework of Dynamic Dual Attention Networks (DDAN) which models individuals' content production strategies through a generative process, under the influence of social interactions involved in the process. Extensive experimental results illustrate the model's effectiveness in user behavior modeling. We make three strong empirical findings: (1) Different strategies give rise to different social payoffs; (2) The best performing individuals exhibit stability in their preference over the discovered strategies, which indicates the emergence of strategic behavior; and (3) The stability of a user's preference is correlated with high payoffs.
View PDF Chat

Leveraging Social Foci for Information Seeking in Social Media

2015-02-09
Suhas Ranganath , Jiliang Tang , Xia Hu +2 more
The rise of social media provides a great opportunity for people to reach out to their social connections to satisfy their information needs. However, generic social media platforms are not … The rise of social media provides a great opportunity for people to reach out to their social connections to satisfy their information needs. However, generic social media platforms are not explicitly designed to assist information seeking of users. In this paper, we propose a novel framework to identify the social connections of a user able to satisfy his information needs. The information need of a social media user is subjective and personal, and we investigate the utility of his social context to identify people able to satisfy it. We present questions users post on Twitter as instances of information seeking activities in social media. We infer soft community memberships of the asker and his social connections by integrating network and content information. Drawing concepts from the social foci theory, we identify answerers whose community memberships in the question domain overlap with that of the asker. Our experiments demonstrate that the framework is effective in identifying answerers to social media questions.
View PDF Chat

How Do We Find Early Adopters Who Will Guide a Resource Constrained Network Towards a Desired Distribution of Behaviors?

2013-01-01
Kaushik Sarkar , Hari Sundaram
We identify influential early adopters that achieve a target behavior distribution for a resource constrained social network with multiple costly behaviors. This problem is important for applications ranging from collective … We identify influential early adopters that achieve a target behavior distribution for a resource constrained social network with multiple costly behaviors. This problem is important for applications ranging from collective behavior change to corporate viral marketing campaigns. In this paper, we propose a model of diffusion of multiple behaviors when individual participants have resource constraints. Individuals adopt the set of behaviors that maximize their utility subject to available resources. We show that the problem of influence maximization for multiple behaviors is NP-complete. Thus we propose heuristics, which are based on node degree and expected immediate adoption, to select early adopters. We evaluate the effectiveness under three metrics: unique number of participants, total number of active behaviors and network resource utilization. We also propose heuristics to distribute the behaviors amongst the early adopters to achieve a target distribution in the population. We test our approach on synthetic and real-world topologies with excellent results. Our heuristics produce 15-51\% increase in resource utilization over the naïve approach.

Influencing Busy People in a Social Network

2016-10-06
Kaushik Sarkar , Hari Sundaram
We identify influential early adopters in a social network, where individuals are resource constrained, to maximize the spread of multiple, costly behaviors. A solution to this problem is especially important … We identify influential early adopters in a social network, where individuals are resource constrained, to maximize the spread of multiple, costly behaviors. A solution to this problem is especially important for viral marketing. The problem of maximizing influence in a social network is challenging since it is computationally intractable. We make three contributions. First, propose a new model of collective behavior that incorporates individual intent, knowledge of neighbors actions and resource constraints. Second, we show that the multiple behavior influence maximization is NP-hard. Furthermore, we show that the problem is submodular, implying the existence of a greedy solution that approximates the optimal solution to within a constant. However, since the greedy algorithm is expensive for large networks, we propose efficient heuristics to identify the influential individuals, including heuristics to assign behaviors to the different early adopters. We test our approach on synthetic and real-world topologies with excellent results. We evaluate the effectiveness under three metrics: unique number of participants, total number of active behaviors and network resource utilization. Our heuristics produce 15-51% increase in expected resource utilization over the naive approach.
View PDF Chat

The Size Conundrum: Why Online Knowledge Markets Can Fail at Scale

2018-04-23
Himel Dev , Chase Geigle , Qingtao Hu +2 more
In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge … In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge market framing allows site operators to reason about market failures, and to design policies to prevent them. Our goal is to provide insights on large-scale knowledge market failures through an interpretable model. We explore a set of interpretable economic production models on a large empirical dataset to analyze the dynamics of content generation in knowledge markets. Amongst these, the Cobb-Douglas model best explains empirical data and provides an intuitive explanation for content generation through the concepts of elasticity and diminishing returns. Content generation depends on user participation and also on how specific types of content (e.g. answers) depends on other types (e.g. questions). We show that these factors of content generation have constant elasticity and a percentage increase in any of the inputs leads to a constant percentage increase in the output. Furthermore, markets exhibit diminishing returns-the marginal output decreases as the input is incrementally increased. Knowledge markets also vary on their returns to scale-the increase in output resulting from a proportionate increase in all inputs. Importantly, many knowledge markets exhibit diseconomies of scale-measures of market health (e.g., the percentage of questions with an accepted answer) decrease as a function of the number of participants. The implications of our work are two-fold: site operators ought to design incentives as a function of system size (number of participants); the market lens should shed insight into complex dependencies amongst different content types and participant actions in general social networks.
View PDF Chat

Leveraging Social Foci for Information Seeking in Social Media

2015-01-01
Suhas Ranganath , Jiliang Tang , Xia Hu +2 more
The rise of social media provides a great opportunity for people to reach out to their social connections to satisfy their information needs. However, generic social media platforms are not … The rise of social media provides a great opportunity for people to reach out to their social connections to satisfy their information needs. However, generic social media platforms are not explicitly designed to assist information seeking of users. In this paper, we propose a novel framework to identify the social connections of a user able to satisfy his information needs. The information need of a social media user is subjective and personal, and we investigate the utility of his social context to identify people able to satisfy it. We present questions users post on Twitter as instances of information seeking activities in social media. We infer soft community memberships of the asker and his social connections by integrating network and content information. Drawing concepts from the social foci theory, we identify answerers who share communities with the asker w.r.t. the question. Our experiments demonstrate that the framework is effective in identifying answerers to social media questions.
View PDF Chat

The Size Conundrum

2018-01-01
Himel Dev , Chase Geigle , Qingtao Hu +2 more
In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge … In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge market framing allows site operators to reason about market failures, and to design policies to prevent them. Our goal is to provide insights on large-scale knowledge market failures through an interpretable model. We explore a set of interpretable economic production models on a large empirical dataset to analyze the dynamics of content generation in knowledge markets. Amongst these, the Cobb-Douglas model best explains empirical data and provides an intuitive explanation for content generation through concepts of elasticity and diminishing returns. Content generation depends on user participation and also on how specific types of content (e.g. answers) depends on other types (e.g. questions). We show that these factors of content generation have constant elasticity---a percentage increase in any of the inputs leads to a constant percentage increase in the output. Furthermore, markets exhibit diminishing returns---the marginal output decreases as the input is incrementally increased. Knowledge markets also vary on their returns to scale---the increase in output resulting from a proportionate increase in all inputs. Importantly, many knowledge markets exhibit diseconomies of scale---measures of market health (e.g., the percentage of questions with an accepted answer) decrease as a function of number of participants. The implications of our work are two-fold: site operators ought to design incentives as a function of system size (number of participants); the market lens should shed insight into complex dependencies amongst different content types and participant actions in general social networks.
View PDF Chat

Report of 2017 NSF Workshop on Multimedia Challenges, Opportunities and Research Roadmaps

2019-01-01
Shih‐Fu Chang , Alexander G. Hauptmann , Louis‐Philippe Morency +7 more
With the transformative technologies and the rapidly changing global R&D landscape, the multimedia and multimodal community is now faced with many new opportunities and uncertainties. With the open source dissemination … With the transformative technologies and the rapidly changing global R&D landscape, the multimedia and multimodal community is now faced with many new opportunities and uncertainties. With the open source dissemination platform and pervasive computing resources, new research results are being discovered at an unprecedented pace. In addition, the rapid exchange and influence of ideas across traditional discipline boundaries have made the emphasis on multimedia multimodal research even more important than before. To seize these opportunities and respond to the challenges, we have organized a workshop to specifically address and brainstorm the challenges, opportunities, and research roadmaps for MM research. The two-day workshop, held on March 30 and 31, 2017 in Washington DC, was sponsored by the Information and Intelligent Systems Division of the National Science Foundation of the United States. Twenty-three (23) invited participants were asked to review and identify research areas in the MM field that are most important over the next 10-15 year timeframe. Important topics were selected through discussion and consensus, and then discussed in depth in breakout groups. Breakout groups reported initial discussion results to the whole group, who continued with further extensive deliberation. For each identified topic, a summary was produced after the workshop to describe the main findings, including the state of the art, challenges, and research roadmaps planned for the next 5, 10, and 15 years in the identified area.

Quantifying Voter Biases in Online Platforms

2019-11-07
Himel Dev , Karrie Karahalios , Hari Sundaram
In content-based online platforms, use of aggregate user feedback (say, the sum of votes) is commonplace as the "gold standard" for measuring content quality. Use of vote aggregates, however, is … In content-based online platforms, use of aggregate user feedback (say, the sum of votes) is commonplace as the "gold standard" for measuring content quality. Use of vote aggregates, however, is at odds with the existing empirical literature, which suggests that voters are susceptible to different biases-reputation (e.g., of the poster), social influence (e.g., votes thus far), and position (e.g., answer position). Our goal is to quantify, in an observational setting, the degree of these biases in online platforms. Specifically, what are the causal effects of different impression signals-such as the reputation of the contributing user, aggregate vote thus far, and position of content-on a participant's vote on content? We adopt an instrumental variable (IV) framework to answer this question. We identify a set of candidate instruments, carefully analyze their validity, and then use the valid instruments to reveal the effects of the impression signals on votes. Our empirical study using log data from Stack Exchange websites shows that the bias estimates from our IV approach differ from the bias estimates from the ordinary least squares (OLS) method. In particular, OLS underestimates reputation bias (1.6-2.2x for gold badges) and position bias (up to 1.9x for the initial position) and overestimates social influence bias (1.8-2.3x for initial votes). The implications of our work include: redesigning user interface to avoid voter biases; making changes to platforms' policy to mitigate voter biases; detecting other forms of biases in online platforms.
View PDF Chat

An Induced Multi-Relational Framework for Answer Selection in Community Question Answer Platforms

2019-01-01
Kanika Narang , Chaoqi Yang , Adit Krishnan +3 more
This paper addresses the question of identifying the best candidate answer to a question on Community Question Answer (CQA) forums. The problem is important because Individuals often visit CQA forums … This paper addresses the question of identifying the best candidate answer to a question on Community Question Answer (CQA) forums. The problem is important because Individuals often visit CQA forums to seek answers to nuanced questions. We develop a novel induced relational graph convolutional network (IR-GCN) framework to address the question. We make three contributions. First, we introduce a modular framework that separates the construction of the graph with the label selection mechanism. We use equivalence relations to induce a graph comprising cliques and identify two label assignment mechanisms---label contrast, label sharing. Then, we show how to encode these assignment mechanisms in GCNs. Second, we show that encoding contrast creates discriminative magnification---enhancing the separation between nodes in the embedding space. Third, we show a surprising result---boosting techniques improve learning over familiar stacking, fusion, or aggregation approaches for neural architectures. We show strong results over the state-of-the-art neural baselines in extensive experiments on 50 StackExchange communities.

Quantifying Voter Biases in Online Platforms: An Instrumental Variable Approach

2019-11-07
Himel Dev , Karrie Karahalios , Hari Sundaram
In content-based online platforms, use of aggregate user feedback (say, the sum of votes) is commonplace as the standard for measuring content quality. Use of vote aggregates, however, is at … In content-based online platforms, use of aggregate user feedback (say, the sum of votes) is commonplace as the standard for measuring content quality. Use of vote aggregates, however, is at odds with the existing empirical literature, which suggests that voters are susceptible to different biases-reputation (e.g., of the poster), social influence (e.g., votes thus far), and position (e.g., answer position). Our goal is to quantify, in an observational setting, the degree of these biases in online platforms. Specifically, what are the causal effects of different impression signals-such as the reputation of the contributing user, aggregate vote thus far, and position of content-on a participant's vote on content? We adopt an instrumental variable (IV) framework to answer this question. We identify a set of candidate instruments, carefully analyze their validity, and then use the valid instruments to reveal the effects of the impression signals on votes. Our empirical study using log data from Stack Exchange websites shows that the bias estimates from our IV approach differ from the bias estimates from the ordinary least squares (OLS) method. In particular, OLS underestimates reputation bias (1.6-2.2x for gold badges) and position bias (up to 1.9x for the initial position) and overestimates social influence bias (1.8-2.3x for initial votes). The implications of our work include: redesigning user interface to avoid voter biases; making changes to platforms' policy to mitigate voter biases; detecting other forms of biases in online platforms.
View PDF Chat

What should I Ask: A Knowledge-driven Approach for Follow-up Questions Generation in Conversational Surveys

2022-01-01
Yubin Ge , Ziang Xiao , Jana Diesner +3 more
Generating follow-up questions on the fly could significantly improve conversational survey quality and user experiences by enabling a more dynamic and personalized survey structure. In this paper, we proposed a … Generating follow-up questions on the fly could significantly improve conversational survey quality and user experiences by enabling a more dynamic and personalized survey structure. In this paper, we proposed a novel task for knowledge-driven follow-up question generation in conversational surveys. We constructed a new human-annotated dataset of human-written follow-up questions with dialogue history and labeled knowledge in the context of conversational surveys. Along with the dataset, we designed and validated a set of reference-free Gricean-inspired evaluation metrics to systematically evaluate the quality of generated follow-up questions. We then propose a two-staged knowledge-driven model for the task, which generates informative and coherent follow-up questions by using knowledge to steer the generation process. The experiments demonstrate that compared to GPT-based baseline models, our two-staged model generates more informative, coherent, and clear follow-up questions.

Finding by Counting

2017-10-06
Subham De , Shreyans Chowdhary , Aniket Shirke +3 more
We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor … We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor localization to play an important role in indoor public spaces in the future. We model the probability of reception of a packet as a generalized quadratic function of distance, beacon power and advertising frequency. Then, we use a Bayesian formulation to determine the coefficients of the packet loss model using empirical observations from our testbed. We develop a new sequential Monte-Carlo algorithm that uses our packet count model. The algorithm is general enough to accommodate different spatial configurations. We have good indoor localization experiments: our approach has an average error of ~ 1.2m, 53% lower than the baseline range-free Monte-Carlo localization algorithm.
View PDF Chat

Social Event Detection with Interaction Graph Modeling

2012-01-01
Yanxiang Wang , Hari Sundaram , Lexing Xie
This paper focuses on detecting social, physical-world events from photos posted on social media sites. The problem is important: cheap media capture devices have significantly increased the number of photos … This paper focuses on detecting social, physical-world events from photos posted on social media sites. The problem is important: cheap media capture devices have significantly increased the number of photos shared on these sites. The main contribution of this paper is to incorporate online social interaction features in the detection of physical events. We believe that online social interaction reflect important signals among the participants on the "social affinity" of two photos, thereby helping event detection. We compute social affinity via a random-walk on a social interaction graph to determine similarity between two photos on the graph. We train a support vector machine classifier to combine the social affinity between photos and photo-centric metadata including time, location, tags and description. Incremental clustering is then used to group photos to event clusters. We have very good results on two large scale real-world datasets: Upcoming and MediaEval. We show an improvement between 0.06-0.10 in F1 on these datasets.
View PDF Chat

Finding by Counting: A Probabilistic Packet Count Model for Indoor Localization in BLE Environments

2017-10-20
Subham De , Shreyans Chowdhary , Aniket Shirke +3 more
We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor … We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor localization to play an important role in indoor public spaces in the future. We model the probability of reception of a packet as a generalized quadratic function of distance, beacon power and advertising frequency. Then, we use a Bayesian formulation to determine the coefficients of the packet loss model using empirical observations from our testbed. We develop a new sequential Monte-Carlo algorithm that uses our packet count model. The algorithm is general enough to accommodate different spatial configurations. We have good indoor localization experiments: our approach has an average error of ~ 1.2m, 53% lower than the baseline range-free Monte-Carlo localization algorithm.
View PDF Chat

Improving Latent User Models in Online Social Media.

2017-11-30
Adit Krishnan , Ashish Sharma , Hari Sundaram
Modern social platforms are characterized by the presence of rich user-behavior data associated with the publication, sharing and consumption of textual content. Users interact with content and with each other … Modern social platforms are characterized by the presence of rich user-behavior data associated with the publication, sharing and consumption of textual content. Users interact with content and with each other in a complex and dynamic social environment while simultaneously evolving over time. In order to effectively characterize users and predict their future behavior in such a setting, it is necessary to overcome several challenges. Content heterogeneity and temporal inconsistency of behavior data result in severe sparsity at the user level. In this paper, we propose a novel mutual-enhancement framework to simultaneously partition and learn latent activity profiles of users. We propose a flexible user partitioning approach to effectively discover rare behaviors and tackle user-level sparsity. We extensively evaluate the proposed framework on massive datasets from real-world platforms including Q&A networks and interactive online courses (MOOCs). Our results indicate significant gains over state-of-the-art behavior models ( 15% avg ) in a varied range of tasks and our gains are further magnified for users with limited interaction data. The proposed algorithms are amenable to parallelization, scale linearly in the size of datasets, and provide flexibility to model diverse facets of user behavior.

Growing Attributed Networks through Local Processes

2019-05-13
Harshay Shah , Suhansanu Kuma , Hari Sundaram
This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and … This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and resource-constrained mechanisms explain the emergence of structural properties found in real-world networks. We make three contributions. First, we propose a simple and accurate model of attributed network growth that jointly explains the emergence of in-degree, local clustering, clustering-degree relationship and attribute mixing patterns. Second, we make use of biased random walks to develop a model that forms edges locally, without recourse to global information. Third, we account for multiple sociological phenomena: bounded rationality; structural constraints; triadic closure; attribute homophily; preferential attachment. Our experiments show that the proposed Attributed Random Walk (ARW) model accurately preserves network structure and attribute mixing patterns of real-world networks; it improves upon the performance of eight well-known models by a significant margin of 2.5-10 × .
View PDF Chat

It's just a matter of perspective(s): Crowd-Powered Consensus Organization of Corpora

2016-01-01
Ayush Jain , Joon Young Seo , Karan Goel +3 more
We study the problem of organizing a collection of objects - images, videos - into clusters, using crowdsourcing. This problem is notoriously hard for computers to do automatically, and even … We study the problem of organizing a collection of objects - images, videos - into clusters, using crowdsourcing. This problem is notoriously hard for computers to do automatically, and even with crowd workers, is challenging to orchestrate: (a) workers may cluster based on different latent hierarchies or perspectives; (b) workers may cluster at different granularities even when clustering using the same perspective; and (c) workers may only see a small portion of the objects when deciding how to cluster them (and therefore have limited understanding of the "big picture"). We develop cost-efficient, accurate algorithms for identifying the consensus organization (i.e., the organizing perspective most workers prefer to employ), and incorporate these algorithms into a cost-effective workflow for organizing a collection of objects, termed ORCHESTRA. We compare our algorithms with other algorithms for clustering, on a variety of real-world datasets, and demonstrate that ORCHESTRA organizes items better and at significantly lower costs.

Task-driven sampling of attributed networks

2016-01-01
Suhansanu Kumar , Hari Sundaram
This paper introduces new techniques for sampling attributed networks to support standard Data Mining tasks. The problem is important for two reasons. First, it is commonplace to perform data mining … This paper introduces new techniques for sampling attributed networks to support standard Data Mining tasks. The problem is important for two reasons. First, it is commonplace to perform data mining tasks such as clustering and classification of network attributes (attributes of the nodes, including social media posts). Furthermore, the extraordinarily large size of real-world networks necessitates that we work with a smaller graph sample. Second, while random sampling will provide an unbiased estimate of content, random access is often unavailable for many networks. Hence, network samplers such as Snowball sampling, Forest Fire, Random Walk, Metropolis-Hastings Random Walk are widely used; however, these attribute-agnostic samplers were designed to capture salient properties of network structure, not node content. The latter is critical for clustering and classification tasks. There are three contributions of this paper. First, we introduce several attribute-aware samplers based on Information Theoretic principles. Second, we prove that these samplers have a bias towards capturing new content, and are equivalent to uniform sampling in the limit. Finally, our experimental results over large real-world datasets and synthetic benchmarks are insightful: attribute-aware samplers outperform both random sampling and baseline attribute-agnostic samplers by a wide margin in clustering and classification tasks.

Modeling Network Growth under Resource Constraints.

2017-12-29
Harshay Shah , Suhansanu Kumar , Hari Sundaram
We propose a resource-constrained network growth model that explains the emergence of key structural properties of real-world directed networks: heavy-tailed indegree distribution, high local clustering and degree-clustering relationship. In real-world … We propose a resource-constrained network growth model that explains the emergence of key structural properties of real-world directed networks: heavy-tailed indegree distribution, high local clustering and degree-clustering relationship. In real-world networks, individuals form edges under constraints of limited network access and partial information. However, well-known growth models that preserve multiple structural properties do not incorporate these resource constraints. Conversely, existing resource-constrained models do not jointly preserve multiple structural properties of real-world networks. We propose a random walk growth model that explains how real-world network properties can jointly arise from edge formation under resource constraints. In our model, each node that joins the network selects a seed node from which it initiates a random walk. At each step of the walk, the new node either jumps back to the seed node or chooses an outgoing or incoming edge to visit another node. It links to each visited node with some probability and stops after forming a few edges. Our experimental results against four well-known growth models indicate improvement in accurately preserving structural properties of five citation networks. Our model also preserves two structural properties that most growth models cannot: skewed local clustering distribution and bivariate indegree-clustering relationship.

Discovering Archetypes to Interpret Evolution of Individual Behavior

2019-01-01
Kanika Narang , Austin Chung , Hari Sundaram +1 more
In this paper, we aim to discover archetypical patterns of individual evolution in large social networks. In our work, an archetype comprises of $\textit{progressive stages}$ of distinct behavior. We introduce … In this paper, we aim to discover archetypical patterns of individual evolution in large social networks. In our work, an archetype comprises of $\textit{progressive stages}$ of distinct behavior. We introduce a novel Gaussian Hidden Markov Model (G-HMM) Cluster to identify archetypes of evolutionary patterns. G-HMMs allow for: near limitless behavioral variation; imposing constraints on how individuals can evolve; different evolutionary rates; and are parsimonious. Our experiments with Academic and StackExchange dataset discover insightful archetypes. We identify four archetypes for researchers: $\textit{Steady}$, $\textit{Diverse, Evolving and Diffuse}$. We observe clear differences in the evolution of male and female researchers within the same archetype. Specifically, women and men differ within an archetype (e.g. Diverse) in how they start, how they transition and the time spent in mid-career. We also found that the differences in grant income are better explained by the differences in archetype than by differences in gender. For StackOverflow, discovered archetypes could be labeled as $\textit{Experts, Seekers, Enthusiasts, and Facilitators}$. We have strong quantitative results with competing baselines for activity prediction and perplexity. For future session prediction, the proposed G-HMM cluster model improves by an average of $32\%$ for different Stack Exchanges and $24\%$ for Academic dataset. Our model also exhibits lower perplexity than the baselines.

Mining Hidden Populations through Attributed Search

2019-01-01
Suhansanu Kumar , Heting Gao , Changyu Wang +2 more
Researchers often query online social platforms through their application programming interfaces (API) to find target populations such as people with mental illness~\cite{De-Choudhury2017} and jazz musicians~\cite{heckathorn2001finding}. Entities of such target population … Researchers often query online social platforms through their application programming interfaces (API) to find target populations such as people with mental illness~\cite{De-Choudhury2017} and jazz musicians~\cite{heckathorn2001finding}. Entities of such target population satisfy a property that is typically identified using an oracle (human or a pre-trained classifier). When the property of the target entities is not directly queryable via the API, we refer to the property as `hidden' and the population as a hidden population. Finding individuals who belong to these populations on social networks is hard because they are non-queryable, and the sampler has to explore from a combinatorial query space within a finite budget limit. By exploiting the correlation between queryable attributes and the population of interest and by hierarchically ordering the query space, we propose a Decision tree-based Thompson sampler (\texttt{DT-TMP}) that efficiently discovers the right combination of attributes to query. Our proposed sampler outperforms the state-of-the-art samplers in online experiments, for example by 54\% on Twitter. When the number of matching entities to a query is known in offline experiments, \texttt{DT-TMP} performs exceedingly well by a factor of 0.9-1.5$\times$ over the baseline samplers. In the future, we wish to explore the option of finding hidden populations by formulating more complex queries.

Growing Attributed Networks through Local Processes.

2017-12-29
Harshay Shah , Suhansanu Kumar , Hari Sundaram
This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and … This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and resource-constrained mechanisms explain the emergence of rich structural properties found in real-world networks. We make three contributions. First, we propose a parsimonious and accurate model of attributed network growth that jointly explains the emergence of in-degree distributions, local clustering, clustering-degree relationship and attribute mixing patterns. Second, our model is based on biased random walks and uses local processes to form edges without recourse to global network information. Third, we account for multiple sociological phenomena: bounded rationality, structural constraints, triadic closure, attribute homophily, and preferential attachment. Our experiments indicate that the proposed Attributed Random Walk (ARW) model accurately preserves network structure and attribute mixing patterns of six real-world networks; it improves upon the performance of eight state-of-the-art models by a statistically significant margin of 2.5-10x.

Mastering the ABCDs of Complex Questions: Answer-Based Claim Decomposition for Fine-grained Self-Evaluation

2023-01-01
Nishant Balepur , Jie Huang , Samraj Moorjani +2 more
When answering complex questions, large language models (LLMs) may produce answers that do not satisfy all criteria of the question. While existing self-evaluation techniques aim to detect if such answers … When answering complex questions, large language models (LLMs) may produce answers that do not satisfy all criteria of the question. While existing self-evaluation techniques aim to detect if such answers are correct, these techniques are unable to determine which criteria of the question are satisfied by the generated answers. To address this issue, we propose answer-based claim decomposition (ABCD), a prompting strategy that decomposes questions into a series of true/false claims that can be used to verify which criteria of the input question an answer satisfies. Using the decomposed ABCD claims, we perform fine-grained self-evaluation. Through preliminary experiments on three datasets, including a newly-collected challenge dataset ObscureQA, we find that GPT-3.5 has some ability to determine to what extent its answer satisfies the criteria of the input question, and can give insights into the errors and knowledge gaps of the model.
View PDF Chat

Audience-Centric Natural Language Generation via Style Infusion

2022-01-01
Samraj Moorjani , Adit Krishnan , Hari Sundaram +2 more
Adopting contextually appropriate, audience-tailored linguistic styles is critical to the success of user-centric language generation systems (e.g., chatbots, computer-aided writing, dialog systems). While existing approaches demonstrate text style transfer (TST) … Adopting contextually appropriate, audience-tailored linguistic styles is critical to the success of user-centric language generation systems (e.g., chatbots, computer-aided writing, dialog systems). While existing approaches demonstrate text style transfer (TST) with large volumes of parallel or non-parallel data, we argue that grounding style on audience-independent external factors is innately limiting for two reasons. First, it is difficult to collect large volumes of audience-specific stylistic data. Second, some stylistic objectives (e.g., persuasiveness, memorability, empathy) are hard to define without audience feedback. In this paper, we propose the novel task of style infusion - infusing the stylistic preferences of audiences in pretrained language generation models. Since humans are better at pairwise comparisons than direct scoring - i.e., is Sample-A more persuasive/polite/empathic than Sample-B - we leverage limited pairwise human judgments to bootstrap a style analysis model and augment our seed set of judgments. We then infuse the learned textual style in a GPT-2 based text generator while balancing fluency and style adoption. With quantitative and qualitative assessments, we show that our infusion approach can generate compelling stylized examples with generic text prompts. We make the anonymized code and data accessible.
View PDF Chat

CodeScope: An Execution-based Multilingual Multitask Multidimensional Benchmark for Evaluating LLMs on Code Understanding and Generation

2023-01-01
Weixiang Yan , Haitian Liu , Yunkun Wang +7 more
Large Language Models (LLMs) have demonstrated remarkable performance on coding related tasks, particularly on assisting humans in programming and facilitating programming automation. However, existing benchmarks for evaluating the code understanding … Large Language Models (LLMs) have demonstrated remarkable performance on coding related tasks, particularly on assisting humans in programming and facilitating programming automation. However, existing benchmarks for evaluating the code understanding and generation capacities of LLMs suffer from severe limitations. First, most benchmarks are deficient as they focus on a narrow range of popular programming languages and specific tasks, whereas the real-world software development scenarios show dire need to implement systems with multilingual programming environments to satisfy diverse requirements. Practical programming practices also strongly expect multi-task settings for testing coding capabilities of LLMs comprehensively and robustly. Second, most benchmarks also fail to consider the actual executability and the consistency of execution results of the generated code. To bridge these gaps between existing benchmarks and expectations from practical applications, we introduce CodeScope, an execution-based, multilingual, multi-task, multi-dimensional evaluation benchmark for comprehensively gauging LLM capabilities on coding tasks. CodeScope covers 43 programming languages and 8 coding tasks. It evaluates the coding performance of LLMs from three dimensions (perspectives): difficulty, efficiency, and length. To facilitate execution-based evaluations of code generation, we develop MultiCodeEngine, an automated code execution engine that supports 14 programming languages. Finally, we systematically evaluate and analyze 8 mainstream LLMs on CodeScope tasks and demonstrate the superior breadth and challenges of CodeScope for evaluating LLMs on code understanding and generation tasks compared to other benchmarks. The CodeScope benchmark and datasets are publicly available at https://github.com/WeixiangYAN/CodeScope.
View PDF Chat

Network Cartography: Seeing the Forest and the Trees

2015-01-01
Jia Wang , Kevin Chen–Chuan Chang , Hari Sundaram
Real-world networks are often complex and large with millions of nodes, posing a great challenge for analysts to quickly see the big picture for more productive subsequent analysis. We aim … Real-world networks are often complex and large with millions of nodes, posing a great challenge for analysts to quickly see the big picture for more productive subsequent analysis. We aim at facilitating exploration of node-attributed networks by creating representations with conciseness, expressiveness, interpretability, and multi-resolution views. We develop such a representation as a {\it map} --- among the first to explore principled network cartography for general networks. In parallel with common maps, ours has landmarks, which aggregate nodes homogeneous in their traits and interactions with nodes elsewhere, and roads, which represent the interactions between the landmarks. We capture such homogeneity by the similar roles the nodes played. Next, to concretely model the landmarks, we propose a probabilistic generative model of networks with roles as latent factors. Furthermore, to enable interactive zooming, we formulate novel model-based constrained optimization. Then, we design efficient linear-time algorithms for the optimizations. Experiments using real-world and synthetic networks show that our method produces more expressive maps than existing methods, with up to 10 times improvement in network reconstruction quality. We also show that our method extracts landmarks with more homogeneous nodes, with up to 90\% improvement in the average attribute/link entropy among the nodes over each landmark. Sense-making of a real-world network using a map computed by our method qualitatively verify the effectiveness of our method.

Finding by Counting: A Probabilistic Packet Count Model for Indoor Localization in BLE Environments

2017-08-27
Subham De , Shreyans Chowdhary , Aniket Shirke +3 more
We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor … We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor localization to play an important role in indoor public spaces in the future. We model the probability of reception of a packet as a generalized quadratic function of distance, beacon power and advertising frequency. Then, we use a Bayesian formulation to determine the coefficients of the packet loss model using empirical observations from our testbed. We develop a new sequential Monte-Carlo algorithm that uses our packet count model. The algorithm is general enough to accommodate different spatial configurations. We have good indoor localization experiments: our approach has an average error of ~1.2m, 53% lower than the baseline range-free Monte-Carlo localization algorithm.
View PDF Chat

The Size Conundrum: Why Online Knowledge Markets Can Fail at Scale

2017-12-01
Himel Dev , Chase Geigle , Qingtao Hu +2 more
In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge … In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge market framing allows site operators to reason about market failures, and to design policies to prevent them. Our goal is to provide insights on large-scale knowledge market failures through an interpretable model. We explore a set of interpretable economic production models on a large empirical dataset to analyze the dynamics of content generation in knowledge markets. Amongst these, the Cobb-Douglas model best explains empirical data and provides an intuitive explanation for content generation through concepts of elasticity and diminishing returns. Content generation depends on user participation and also on how specific types of content (e.g. answers) depends on other types (e.g. questions). We show that these factors of content generation have constant elasticity---a percentage increase in any of the inputs leads to a constant percentage increase in the output. Furthermore, markets exhibit diminishing returns---the marginal output decreases as the input is incrementally increased. Knowledge markets also vary on their returns to scale---the increase in output resulting from a proportionate increase in all inputs. Importantly, many knowledge markets exhibit diseconomies of scale---measures of market health (e.g., the percentage of questions with an accepted answer) decrease as a function of number of participants. The implications of our work are two-fold: site operators ought to design incentives as a function of system size (number of participants); the market lens should shed insight into complex dependencies amongst different content types and participant actions in general social networks.
View PDF Chat

Resource Allocation Game on Social Networks: Best Response Dynamics and Convergence

2018-01-01
Wei-Chun Lee , Vasilis Livanos , Ruta Mehta +1 more
The decisions that human beings make to allocate time has significant bearing on economic output and to the sustenance of social networks. The time allocation problem motivates our formal analysis … The decisions that human beings make to allocate time has significant bearing on economic output and to the sustenance of social networks. The time allocation problem motivates our formal analysis of the resource allocation game, where agents on a social network, who have asymmetric, private interaction preferences, make decisions on how to allocate time, a bounded endowment, over their neighbors. Unlike the well-known opinion formation game on a social network, our game appears not to be a potential game, and the Best-Response dynamics is non-differentiable making the analysis of Best-Response dynamics non-trivial. In our game, we consider two types of player behavior, namely optimistic or pessimistic, based on how they use their time endowment over their neighbors. To analyze Best-Response dynamics, we circumvent the problem of the game not being a potential game, through the lens of a novel two-level potential function approach. We show that the Best-Response dynamics converges point-wise to a Nash Equilibrium when players are all: optimistic; pessimistic; a mix of both types. Finally, we show that the Nash Equilibrium set is non-convex but connected, and Price of Anarchy is unbounded while Price of Stability is one. Extensive simulations over a stylized grid reveals that the distribution of quality of the convergence points is unimodal-we conjecture that presence of unimodality is tied to the connectedness of Nash Equilibrium.

Growing Attributed Networks through Local Processes

2017-01-01
Harshay Shah , Suhansanu Kumar , Hari Sundaram
This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and … This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and resource-constrained mechanisms explain the emergence of rich structural properties found in real-world networks. We make three contributions. First, we propose a parsimonious and accurate model of attributed network growth that jointly explains the emergence of in-degree distributions, local clustering, clustering-degree relationship and attribute mixing patterns. Second, our model is based on biased random walks and uses local processes to form edges without recourse to global network information. Third, we account for multiple sociological phenomena: bounded rationality, structural constraints, triadic closure, attribute homophily, and preferential attachment. Our experiments indicate that the proposed Attributed Random Walk (ARW) model accurately preserves network structure and attribute mixing patterns of six real-world networks; it improves upon the performance of eight state-of-the-art models by a statistically significant margin of 2.5-10x.
View PDF Chat

Improving Latent User Models in Online Social Media

2017-01-01
Adit Krishnan , Ashish Sharma , Hari Sundaram
Modern social platforms are characterized by the presence of rich user-behavior data associated with the publication, sharing and consumption of textual content. Users interact with content and with each other … Modern social platforms are characterized by the presence of rich user-behavior data associated with the publication, sharing and consumption of textual content. Users interact with content and with each other in a complex and dynamic social environment while simultaneously evolving over time. In order to effectively characterize users and predict their future behavior in such a setting, it is necessary to overcome several challenges. Content heterogeneity and temporal inconsistency of behavior data result in severe sparsity at the user level. In this paper, we propose a novel mutual-enhancement framework to simultaneously partition and learn latent activity profiles of users. We propose a flexible user partitioning approach to effectively discover rare behaviors and tackle user-level sparsity. We extensively evaluate the proposed framework on massive datasets from real-world platforms including Q&A networks and interactive online courses (MOOCs). Our results indicate significant gains over state-of-the-art behavior models ( 15% avg ) in a varied range of tasks and our gains are further magnified for users with limited interaction data. The proposed algorithms are amenable to parallelization, scale linearly in the size of datasets, and provide flexibility to model diverse facets of user behavior.
View PDF Chat

Audience-Centric Natural Language Generation via Style Infusion

2023-01-01
Samraj Moorjani , Adit Krishnan , Hari Sundaram +2 more
Adopting contextually appropriate, audience-tailored linguistic styles is critical to the success of user-centric language generation systems (e.g., chatbots, computer-aided writing, dialog systems). While existing approaches demonstrate textual style transfer with … Adopting contextually appropriate, audience-tailored linguistic styles is critical to the success of user-centric language generation systems (e.g., chatbots, computer-aided writing, dialog systems). While existing approaches demonstrate textual style transfer with large volumes of parallel or non-parallel data, we argue that grounding style on audience-independent external factors is innately limiting for two reasons. First, it is difficult to collect large volumes of audience-specific stylistic data. Second, some stylistic objectives (e.g., persuasiveness, memorability, empathy) are hard to define without audience feedback. In this paper, we propose the novel task of style infusion - infusing the stylistic preferences of audiences in pretrained language generation models. Since humans are better at pairwise comparisons than direct scoring - i.e., is Sample-A more persuasive/polite/empathic than Sample-B - we leverage limited pairwise human judgments to bootstrap a style analysis model and augment our seed set of judgments. We then infuse the learned textual style in a GPT-2 based text generator while balancing fluency and style adoption. With quantitative and qualitative assessments, we show that our infusion approach can generate compelling stylized examples with generic text prompts. The code and data are accessible at https://github.com/CrowdDynamicsLab/StyleInfusion.
View PDF Chat

Enhancing Generation through Summarization Duality and Explicit Outline Control

2023-01-01
Yunzhe Li , Qian Chen , Weixiang Yan +3 more
Automatically open-ended long text generation poses significant challenges due to semantic incoherence and plot implausibility. Previous works usually alleviate this problem through outlines in the form of short phrases or … Automatically open-ended long text generation poses significant challenges due to semantic incoherence and plot implausibility. Previous works usually alleviate this problem through outlines in the form of short phrases or abstractive signals by designing unsupervised tasks, which tend to be unstable and weakly interpretable. Assuming that a summary serves as a mature outline, we introduce a two-stage, summary-enhanced outline supervised generation framework. This framework leverages the dual characteristics of the summarization task to improve outline prediction, resulting in more explicit and plausible outlines. Furthermore, we identify an underutilization issue in outline-based generation with both standard pretrained language models (e.g., GPT-2, BART) and large language models (e.g., Vicuna, ChatGPT). To address this, we propose a novel explicit outline control method for more effective utilization of generated outlines.

Packet Reception Probability: Packets That You Can't Decode Can Help Keep You Safe

2023-01-01
Subham De , Deepak Vasisht , Hari Sundaram +1 more
This paper provides a robust, scalable Bluetooth Low-Energy (BLE) based indoor localization solution using commodity hardware. While WiFi-based indoor localization has been widely studied, BLE has emerged a key technology … This paper provides a robust, scalable Bluetooth Low-Energy (BLE) based indoor localization solution using commodity hardware. While WiFi-based indoor localization has been widely studied, BLE has emerged a key technology for contact-tracing in the current pandemic. To accurately estimate distance using BLE on commercial devices, systems today rely on Receiver Signal Strength Indicator(RSSI) which suffers from sampling bias and multipath effects. We propose a new metric: Packet Reception Probability (PRP) that builds on a counter-intuitive idea that we can exploit packet loss to estimate distance. We localize using a Bayesian-PRP formulation that also incorporates an explicit model of the multipath. To make deployment easy, we do not require any hardware, firmware, or driver-level changes to off-the-shelf devices, and require minimal training. PRP can achieve meter level accuracy with just 6 devices with known locations and 12 training locations. We show that fusing PRP with RSSI is beneficial at short distances < 2m. Beyond 2m, fusion is worse than PRP, as RSSI becomes effectively de-correlated with distance. Robust location accuracy at all distances and ease of deployment with PRP can help enable wide range indoor localization solutions using BLE.
View PDF Chat

Pre-trained Neural Recommenders: A Transferable Zero-Shot Framework for Recommendation Systems

2023-01-01
Junting Wang , Adit Krishnan , Hari Sundaram +1 more
Modern neural collaborative filtering techniques are critical to the success of e-commerce, social media, and content-sharing platforms. However, despite technical advances -- for every new application domain, we need to … Modern neural collaborative filtering techniques are critical to the success of e-commerce, social media, and content-sharing platforms. However, despite technical advances -- for every new application domain, we need to train an NCF model from scratch. In contrast, pre-trained vision and language models are routinely applied to diverse applications directly (zero-shot) or with limited fine-tuning. Inspired by the impact of pre-trained models, we explore the possibility of pre-trained recommender models that support building recommender systems in new domains, with minimal or no retraining, without the use of any auxiliary user or item information. Zero-shot recommendation without auxiliary information is challenging because we cannot form associations between users and items across datasets when there are no overlapping users or items. Our fundamental insight is that the statistical characteristics of the user-item interaction matrix are universally available across different domains and datasets. Thus, we use the statistical characteristics of the user-item interaction matrix to identify dataset-independent representations for users and items. We show how to learn universal (i.e., supporting zero-shot adaptation without user or item auxiliary information) representations for nodes and edges from the bipartite user-item interaction graph. We learn representations by exploiting the statistical properties of the interaction data, including user and item marginals, and the size and density distributions of their clusters.

Friends with Costs and Benefits: Community Formation with Myopic, Boundedly-Rational Actors

2023-01-01
Naina Balepur , Andy Lee , Hari Sundaram
In this paper we address how complex social communities emerge from local decisions by individuals with limited attention and knowledge. This problem is critical; if we understand community formation mechanisms, … In this paper we address how complex social communities emerge from local decisions by individuals with limited attention and knowledge. This problem is critical; if we understand community formation mechanisms, it may be possible to intervene to improve social welfare. We propose an interpretable, novel model for attributed community formation driven by resource-bounded individuals' strategic, selfish behavior. In our stylized model, attributed individuals act strategically in two dimensions: attribute and network structure. Agents are endowed with limited attention, and communication costs limit the number of active connections. In each time step, each agent proposes a new friendship. Agents then accept proposals, decline proposals, or remove friends, consistent with their strategy to maximize payoff. We identify criteria (number of stable triads) for convergence to some community structure and prove that our community formation model converges to a stable network. Ablations justify the ecological validity of our model and show that each aspect of the model is essential. Our empirical results on a physical world microfinance community demonstrate excellent model fits compared to baseline models.

A Pre-trained Sequential Recommendation Framework: Popularity Dynamics for Zero-shot Transfer

2024-01-01
Junting Wang , Praneet Rathi , Hari Sundaram
Sequential recommenders are crucial to the success of online applications, \eg e-commerce, video streaming, and social media. While model architectures continue to improve, for every new application domain, we still … Sequential recommenders are crucial to the success of online applications, \eg e-commerce, video streaming, and social media. While model architectures continue to improve, for every new application domain, we still have to train a new model from scratch for high quality recommendations. On the other hand, pre-trained language and vision models have shown great success in zero-shot or few-shot adaptation to new application domains. Inspired by the success of pre-trained models in peer AI fields, we propose a novel pre-trained sequential recommendation framework: PrepRec. We learn universal item representations by modeling item popularity dynamics. Through extensive experiments on five real-world datasets, we show that PrepRec, without any auxiliary information, can not only zero-shot transfer to a new domain, but achieve competitive performance compared to state-of-the-art sequential recommender models with only a fraction of the model size. In addition, with a simple post-hoc interpolation, PrepRec can improve the performance of existing sequential recommenders on average by 13.8\% in Recall@10 and 29.5% in NDCG@10. We provide an anonymized implementation of PrepRec at https://anonymous.4open.science/r/PrepRec--2F60/
View PDF Chat

CEV-LM: Controlled Edit Vector Language Model for Shaping Natural Language Generations

2024-02-22
Samraj Moorjani , Adit Krishnan , Hari Sundaram
As large-scale language models become the standard for text generation, there is a greater need to tailor the generations to be more or less concise, targeted, and informative, depending on … As large-scale language models become the standard for text generation, there is a greater need to tailor the generations to be more or less concise, targeted, and informative, depending on the audience/application. Existing control approaches primarily adjust the semantic (e.g., emotion, topics), structural (e.g., syntax tree, parts-of-speech), and lexical (e.g., keyword/phrase inclusion) properties of text, but are insufficient to accomplish complex objectives such as pacing which control the complexity and readability of the text. In this paper, we introduce CEV-LM - a lightweight, semi-autoregressive language model that utilizes constrained edit vectors to control three complementary metrics (speed, volume, and circuitousness) that quantify the shape of text (e.g., pacing of content). We study an extensive set of state-of-the-art CTG models and find that CEV-LM provides significantly more targeted and precise control of these three metrics while preserving semantic content, using less training data, and containing fewer parameters.
View PDF Chat

Spectrum Extraction and Clipping for Implicitly Linear Layers

2024-02-25
Ali Ebrahimpour Boroojeny , Matus Telgarsky , Hari Sundaram
We show the effectiveness of automatic differentiation in efficiently and correctly computing and controlling the spectrum of implicitly linear operators, a rich family of layer types including all standard convolutional … We show the effectiveness of automatic differentiation in efficiently and correctly computing and controlling the spectrum of implicitly linear operators, a rich family of layer types including all standard convolutional and dense layers. We provide the first clipping method which is correct for general convolution layers, and illuminate the representational limitation that caused correctness issues in prior work. We study the effect of the batch normalization layers when concatenated with convolutional layers and show how our clipping method can be applied to their composition. By comparing the accuracy and performance of our algorithms to the state-of-the-art methods, using various experiments, we show they are more precise and efficient and lead to better generalization and adversarial robustness. We provide the code for using our methods at https://github.com/Ali-E/FastClip.
View PDF Chat

Designing Fair Systems for Consumers to Exploit Personalized Pricing

2024-09-04
Aditya Karan , Naina Balepur , Hari Sundaram
Many online marketplaces personalize prices based on consumer attributes. Since these prices are private, consumers will not realize if they spend more on a good than the lowest possible price, … Many online marketplaces personalize prices based on consumer attributes. Since these prices are private, consumers will not realize if they spend more on a good than the lowest possible price, and cannot easily take action to get better prices. In this paper we introduce a system that takes advantage of personalized pricing so consumers can profit while improving fairness. Our system matches consumers for trading; the lower-paying consumer buys the good for the higher-paying consumer for some fee. We explore various modeling choices and fairness targets to determine which schema will leave consumers best off, while also earning revenue for the system itself. We show that when consumers individually negotiate the transaction price, they are able to achieve the most fair outcomes. Conversely, when transaction prices are centrally set, consumers are often unwilling to transact. Minimizing the average price paid by an individual or group is most profitable for the system, while achieving a $67\%$ reduction in prices. We see that a high dispersion (or range) of original prices is necessary for our system to be viable. Higher dispersion can actually lead to increased consumer welfare, and act as a check against extreme personalization. Our results provide theoretical evidence that such a system could improve fairness for consumers while sustaining itself financially.
View PDF Chat

A Pre-trained Zero-shot Sequential Recommendation Framework via Popularity Dynamics

2024-10-08
J. Wang , Praneet Rathi , Hari Sundaram
This paper proposes a novel pre-trained framework for zero-shot cross-domain sequential recommendation without auxiliary information. While using auxiliary information (e.g., item descriptions) seems promising for cross-domain transfer, a cross-domain adaptation … This paper proposes a novel pre-trained framework for zero-shot cross-domain sequential recommendation without auxiliary information. While using auxiliary information (e.g., item descriptions) seems promising for cross-domain transfer, a cross-domain adaptation of sequential recommenders can be challenging when the target domain differs from the source domain—item descriptions are in different languages; metadata modalities (e.g., audio, image, and text) differ across source and target domains. If we can learn universal item representations independent of the domain type (e.g., groceries, movies), we can achieve zero-shot cross-domain transfer without auxiliary information. Our critical insight is that user interaction sequences highlight shifting user preferences via the popularity dynamics of interacted items. We present a pre-trained sequential recommendation framework: PrepRec, which utilizes a novel popularity dynamics-aware transformer architecture. Through extensive experiments on five real-world datasets, we show that PrepRec, without any auxiliary information, can zero-shot adapt to new application domains and achieve competitive performance compared to state-of-the-art sequential recommender models. In addition, we show that PrepRec complements existing sequential recommenders. With a simple post-hoc interpolation, PrepRec improves the performance of existing sequential recommenders on average by 11.8% in Recall@10 and 22% in NDCG@10. We provide an anonymized implementation of PrepRec at https://github.com/CrowdDynamicsLab/preprec.
View PDF Chat

LOTOS: Layer-wise Orthogonalization for Training Robust Ensembles

2024-10-07
Ali Ebrahimpour-Boroojeny , Hari Sundaram , Varun Chandrasekaran
Transferability of adversarial examples is a well-known property that endangers all classification models, even those that are only accessible through black-box queries. Prior work has shown that an ensemble of … Transferability of adversarial examples is a well-known property that endangers all classification models, even those that are only accessible through black-box queries. Prior work has shown that an ensemble of models is more resilient to transferability: the probability that an adversarial example is effective against most models of the ensemble is low. Thus, most ongoing research focuses on improving ensemble diversity. Another line of prior work has shown that Lipschitz continuity of the models can make models more robust since it limits how a model's output changes with small input perturbations. In this paper, we study the effect of Lipschitz continuity on transferability rates. We show that although a lower Lipschitz constant increases the robustness of a single model, it is not as beneficial in training robust ensembles as it increases the transferability rate of adversarial examples across models in the ensemble. Therefore, we introduce LOTOS, a new training paradigm for ensembles, which counteracts this adverse effect. It does so by promoting orthogonality among the top-$k$ sub-spaces of the transformations of the corresponding affine layers of any pair of models in the ensemble. We theoretically show that $k$ does not need to be large for convolutional layers, which makes the computational overhead negligible. Through various experiments, we show LOTOS increases the robust accuracy of ensembles of ResNet-18 models by $6$ percentage points (p.p) against black-box attacks on CIFAR-10. It is also capable of combining with the robustness of prior state-of-the-art methods for training robust ensembles to enhance their robust accuracy by $10.7$ p.p.
View PDF Chat

A Zero-Shot Generalization Framework for LLM-Driven Cross-Domain Sequential Recommendation

2025-01-31
Yunzhe Li , Junting Wang , Hari Sundaram +1 more
Zero-shot cross-domain sequential recommendation (ZCDSR) enables predictions in unseen domains without the need for additional training or fine-tuning, making it particularly valuable in data-sparse environments where traditional models struggle. Recent … Zero-shot cross-domain sequential recommendation (ZCDSR) enables predictions in unseen domains without the need for additional training or fine-tuning, making it particularly valuable in data-sparse environments where traditional models struggle. Recent advancements in large language models (LLMs) have greatly improved ZCDSR by leveraging rich pretrained representations to facilitate cross-domain knowledge transfer. However, a key challenge persists: domain semantic bias, which arises from variations in vocabulary and content focus across domains. This misalignment leads to inconsistencies in item embeddings and hinders generalization. To address this issue, we propose a novel framework designed to enhance LLM-based ZCDSR by improving cross-domain alignment at both the item and sequential levels. At the item level, we introduce a generalization loss that promotes inter-domain compactness by aligning embeddings of similar items across domains while maintaining intra-domain diversity to preserve unique item characteristics. This prevents embeddings from becoming overly generic while ensuring effective transferability. At the sequential level, we develop a method for transferring user behavioral patterns by clustering user sequences in the source domain and applying attention-based aggregation for target domain inference. This dynamic adaptation of user embeddings allows effective zero-shot recommendations without requiring target-domain interactions. Comprehensive experiments across multiple datasets and domains demonstrate that our framework significantly improves sequential recommendation performance in the ZCDSR setting. By mitigating domain bias and enhancing the transferability of sequential patterns, our method provides a scalable and robust approach for achieving more effective zero-shot recommendations across domains.
View PDF Chat

Venire: A Machine Learning-Guided Panel Review System for Community Content Moderation

2024-10-30
Vinay Koshy , Frederick D. S. Choi , Yi-Shyuan Chiang +3 more
Research into community content moderation often assumes that moderation teams govern with a single, unified voice. However, recent work has found that moderators disagree with one another at modest, but … Research into community content moderation often assumes that moderation teams govern with a single, unified voice. However, recent work has found that moderators disagree with one another at modest, but concerning rates. The problem is not the root disagreements themselves. Subjectivity in moderation is unavoidable, and there are clear benefits to including diverse perspectives within a moderation team. Instead, the crux of the issue is that, due to resource constraints, moderation decisions end up being made by individual decision-makers. The result is decision-making that is inconsistent, which is frustrating for community members. To address this, we develop Venire, an ML-backed system for panel review on Reddit. Venire uses a machine learning model trained on log data to identify the cases where moderators are most likely to disagree. Venire fast-tracks these cases for multi-person review. Ideally, Venire allows moderators to surface and resolve disagreements that would have otherwise gone unnoticed. We conduct three studies through which we design and evaluate Venire: a set of formative interviews with moderators, technical evaluations on two datasets, and a think-aloud study in which moderators used Venire to make decisions on real moderation cases. Quantitatively, we demonstrate that Venire is able to improve decision consistency and surface latent disagreements. Qualitatively, we find that Venire helps moderators resolve difficult moderation cases more confidently. Venire represents a novel paradigm for human-AI content moderation, and shifts the conversation from replacing human decision-making to supporting it.
View PDF Chat

A Pre-trained Zero-shot Sequential Recommendation Framework via Popularity Dynamics

2024-10-08
J. Wang , Praneet Rathi , Hari Sundaram
This paper proposes a novel pre-trained framework for zero-shot cross-domain sequential recommendation without auxiliary information. While using auxiliary information (e.g., item descriptions) seems promising for cross-domain transfer, a cross-domain adaptation … This paper proposes a novel pre-trained framework for zero-shot cross-domain sequential recommendation without auxiliary information. While using auxiliary information (e.g., item descriptions) seems promising for cross-domain transfer, a cross-domain adaptation of sequential recommenders can be challenging when the target domain differs from the source domain—item descriptions are in different languages; metadata modalities (e.g., audio, image, and text) differ across source and target domains. If we can learn universal item representations independent of the domain type (e.g., groceries, movies), we can achieve zero-shot cross-domain transfer without auxiliary information. Our critical insight is that user interaction sequences highlight shifting user preferences via the popularity dynamics of interacted items. We present a pre-trained sequential recommendation framework: PrepRec, which utilizes a novel popularity dynamics-aware transformer architecture. Through extensive experiments on five real-world datasets, we show that PrepRec, without any auxiliary information, can zero-shot adapt to new application domains and achieve competitive performance compared to state-of-the-art sequential recommender models. In addition, we show that PrepRec complements existing sequential recommenders. With a simple post-hoc interpolation, PrepRec improves the performance of existing sequential recommenders on average by 11.8% in Recall@10 and 22% in NDCG@10. We provide an anonymized implementation of PrepRec at https://github.com/CrowdDynamicsLab/preprec.
View PDF Chat

LOTOS: Layer-wise Orthogonalization for Training Robust Ensembles

2024-10-07
Ali Ebrahimpour-Boroojeny , Hari Sundaram , Varun Chandrasekaran
Transferability of adversarial examples is a well-known property that endangers all classification models, even those that are only accessible through black-box queries. Prior work has shown that an ensemble of … Transferability of adversarial examples is a well-known property that endangers all classification models, even those that are only accessible through black-box queries. Prior work has shown that an ensemble of models is more resilient to transferability: the probability that an adversarial example is effective against most models of the ensemble is low. Thus, most ongoing research focuses on improving ensemble diversity. Another line of prior work has shown that Lipschitz continuity of the models can make models more robust since it limits how a model's output changes with small input perturbations. In this paper, we study the effect of Lipschitz continuity on transferability rates. We show that although a lower Lipschitz constant increases the robustness of a single model, it is not as beneficial in training robust ensembles as it increases the transferability rate of adversarial examples across models in the ensemble. Therefore, we introduce LOTOS, a new training paradigm for ensembles, which counteracts this adverse effect. It does so by promoting orthogonality among the top-$k$ sub-spaces of the transformations of the corresponding affine layers of any pair of models in the ensemble. We theoretically show that $k$ does not need to be large for convolutional layers, which makes the computational overhead negligible. Through various experiments, we show LOTOS increases the robust accuracy of ensembles of ResNet-18 models by $6$ percentage points (p.p) against black-box attacks on CIFAR-10. It is also capable of combining with the robustness of prior state-of-the-art methods for training robust ensembles to enhance their robust accuracy by $10.7$ p.p.
View PDF Chat

Designing Fair Systems for Consumers to Exploit Personalized Pricing

2024-09-04
Aditya Karan , Naina Balepur , Hari Sundaram
Many online marketplaces personalize prices based on consumer attributes. Since these prices are private, consumers will not realize if they spend more on a good than the lowest possible price, … Many online marketplaces personalize prices based on consumer attributes. Since these prices are private, consumers will not realize if they spend more on a good than the lowest possible price, and cannot easily take action to get better prices. In this paper we introduce a system that takes advantage of personalized pricing so consumers can profit while improving fairness. Our system matches consumers for trading; the lower-paying consumer buys the good for the higher-paying consumer for some fee. We explore various modeling choices and fairness targets to determine which schema will leave consumers best off, while also earning revenue for the system itself. We show that when consumers individually negotiate the transaction price, they are able to achieve the most fair outcomes. Conversely, when transaction prices are centrally set, consumers are often unwilling to transact. Minimizing the average price paid by an individual or group is most profitable for the system, while achieving a $67\%$ reduction in prices. We see that a high dispersion (or range) of original prices is necessary for our system to be viable. Higher dispersion can actually lead to increased consumer welfare, and act as a check against extreme personalization. Our results provide theoretical evidence that such a system could improve fairness for consumers while sustaining itself financially.
View PDF Chat

Spectrum Extraction and Clipping for Implicitly Linear Layers

2024-02-25
Ali Ebrahimpour Boroojeny , Matus Telgarsky , Hari Sundaram
We show the effectiveness of automatic differentiation in efficiently and correctly computing and controlling the spectrum of implicitly linear operators, a rich family of layer types including all standard convolutional … We show the effectiveness of automatic differentiation in efficiently and correctly computing and controlling the spectrum of implicitly linear operators, a rich family of layer types including all standard convolutional and dense layers. We provide the first clipping method which is correct for general convolution layers, and illuminate the representational limitation that caused correctness issues in prior work. We study the effect of the batch normalization layers when concatenated with convolutional layers and show how our clipping method can be applied to their composition. By comparing the accuracy and performance of our algorithms to the state-of-the-art methods, using various experiments, we show they are more precise and efficient and lead to better generalization and adversarial robustness. We provide the code for using our methods at https://github.com/Ali-E/FastClip.
View PDF Chat

CEV-LM: Controlled Edit Vector Language Model for Shaping Natural Language Generations

2024-02-22
Samraj Moorjani , Adit Krishnan , Hari Sundaram
As large-scale language models become the standard for text generation, there is a greater need to tailor the generations to be more or less concise, targeted, and informative, depending on … As large-scale language models become the standard for text generation, there is a greater need to tailor the generations to be more or less concise, targeted, and informative, depending on the audience/application. Existing control approaches primarily adjust the semantic (e.g., emotion, topics), structural (e.g., syntax tree, parts-of-speech), and lexical (e.g., keyword/phrase inclusion) properties of text, but are insufficient to accomplish complex objectives such as pacing which control the complexity and readability of the text. In this paper, we introduce CEV-LM - a lightweight, semi-autoregressive language model that utilizes constrained edit vectors to control three complementary metrics (speed, volume, and circuitousness) that quantify the shape of text (e.g., pacing of content). We study an extensive set of state-of-the-art CTG models and find that CEV-LM provides significantly more targeted and precise control of these three metrics while preserving semantic content, using less training data, and containing fewer parameters.
View PDF Chat

A Pre-trained Sequential Recommendation Framework: Popularity Dynamics for Zero-shot Transfer

2024-01-01
Junting Wang , Praneet Rathi , Hari Sundaram
Sequential recommenders are crucial to the success of online applications, \eg e-commerce, video streaming, and social media. While model architectures continue to improve, for every new application domain, we still … Sequential recommenders are crucial to the success of online applications, \eg e-commerce, video streaming, and social media. While model architectures continue to improve, for every new application domain, we still have to train a new model from scratch for high quality recommendations. On the other hand, pre-trained language and vision models have shown great success in zero-shot or few-shot adaptation to new application domains. Inspired by the success of pre-trained models in peer AI fields, we propose a novel pre-trained sequential recommendation framework: PrepRec. We learn universal item representations by modeling item popularity dynamics. Through extensive experiments on five real-world datasets, we show that PrepRec, without any auxiliary information, can not only zero-shot transfer to a new domain, but achieve competitive performance compared to state-of-the-art sequential recommender models with only a fraction of the model size. In addition, with a simple post-hoc interpolation, PrepRec can improve the performance of existing sequential recommenders on average by 13.8\% in Recall@10 and 29.5% in NDCG@10. We provide an anonymized implementation of PrepRec at https://anonymous.4open.science/r/PrepRec--2F60/
View PDF Chat

Inform the Uninformed: Improving Online Informed Consent Reading with an AI-Powered Chatbot

2023-04-19
Ziang Xiao , Tiffany Wenting Li , Karrie Karahalios +1 more
Informed consent is a core cornerstone of ethics in human subject research. Through the informed consent process, participants learn about the study procedure, benefits, risks, and more to make an … Informed consent is a core cornerstone of ethics in human subject research. Through the informed consent process, participants learn about the study procedure, benefits, risks, and more to make an informed decision. However, recent studies showed that current practices might lead to uninformed decisions and expose participants to unknown risks, especially in online studies. Without the researcher's presence and guidance, online participants must read a lengthy form on their own with no answers to their questions. In this paper, we examined the role of an AI-powered chatbot in improving informed consent online. By comparing the chatbot with form-based interaction, we found the chatbot improved consent form reading, promoted participants' feelings of agency, and closed the power gap between the participant and the researcher. Our exploratory analysis further revealed the altered power dynamic might eventually benefit study response quality. We discussed design implications for creating AI-powered chatbots to offer effective informed consent in broader settings.
View PDF Chat

Audience-Centric Natural Language Generation via Style Infusion

2023-01-01
Samraj Moorjani , Adit Krishnan , Hari Sundaram +2 more
Adopting contextually appropriate, audience-tailored linguistic styles is critical to the success of user-centric language generation systems (e.g., chatbots, computer-aided writing, dialog systems). While existing approaches demonstrate textual style transfer with … Adopting contextually appropriate, audience-tailored linguistic styles is critical to the success of user-centric language generation systems (e.g., chatbots, computer-aided writing, dialog systems). While existing approaches demonstrate textual style transfer with large volumes of parallel or non-parallel data, we argue that grounding style on audience-independent external factors is innately limiting for two reasons. First, it is difficult to collect large volumes of audience-specific stylistic data. Second, some stylistic objectives (e.g., persuasiveness, memorability, empathy) are hard to define without audience feedback. In this paper, we propose the novel task of style infusion - infusing the stylistic preferences of audiences in pretrained language generation models. Since humans are better at pairwise comparisons than direct scoring - i.e., is Sample-A more persuasive/polite/empathic than Sample-B - we leverage limited pairwise human judgments to bootstrap a style analysis model and augment our seed set of judgments. We then infuse the learned textual style in a GPT-2 based text generator while balancing fluency and style adoption. With quantitative and qualitative assessments, we show that our infusion approach can generate compelling stylized examples with generic text prompts. The code and data are accessible at https://github.com/CrowdDynamicsLab/StyleInfusion.
View PDF Chat

Enhancing Generation through Summarization Duality and Explicit Outline Control

2023-01-01
Yunzhe Li , Qian Chen , Weixiang Yan +3 more
Automatically open-ended long text generation poses significant challenges due to semantic incoherence and plot implausibility. Previous works usually alleviate this problem through outlines in the form of short phrases or … Automatically open-ended long text generation poses significant challenges due to semantic incoherence and plot implausibility. Previous works usually alleviate this problem through outlines in the form of short phrases or abstractive signals by designing unsupervised tasks, which tend to be unstable and weakly interpretable. Assuming that a summary serves as a mature outline, we introduce a two-stage, summary-enhanced outline supervised generation framework. This framework leverages the dual characteristics of the summarization task to improve outline prediction, resulting in more explicit and plausible outlines. Furthermore, we identify an underutilization issue in outline-based generation with both standard pretrained language models (e.g., GPT-2, BART) and large language models (e.g., Vicuna, ChatGPT). To address this, we propose a novel explicit outline control method for more effective utilization of generated outlines.

Mastering the ABCDs of Complex Questions: Answer-Based Claim Decomposition for Fine-grained Self-Evaluation

2023-01-01
Nishant Balepur , Jie Huang , Samraj Moorjani +2 more
When answering complex questions, large language models (LLMs) may produce answers that do not satisfy all criteria of the question. While existing self-evaluation techniques aim to detect if such answers … When answering complex questions, large language models (LLMs) may produce answers that do not satisfy all criteria of the question. While existing self-evaluation techniques aim to detect if such answers are correct, these techniques are unable to determine which criteria of the question are satisfied by the generated answers. To address this issue, we propose answer-based claim decomposition (ABCD), a prompting strategy that decomposes questions into a series of true/false claims that can be used to verify which criteria of the input question an answer satisfies. Using the decomposed ABCD claims, we perform fine-grained self-evaluation. Through preliminary experiments on three datasets, including a newly-collected challenge dataset ObscureQA, we find that GPT-3.5 has some ability to determine to what extent its answer satisfies the criteria of the input question, and can give insights into the errors and knowledge gaps of the model.
View PDF Chat

Packet Reception Probability: Packets That You Can't Decode Can Help Keep You Safe

2023-01-01
Subham De , Deepak Vasisht , Hari Sundaram +1 more
This paper provides a robust, scalable Bluetooth Low-Energy (BLE) based indoor localization solution using commodity hardware. While WiFi-based indoor localization has been widely studied, BLE has emerged a key technology … This paper provides a robust, scalable Bluetooth Low-Energy (BLE) based indoor localization solution using commodity hardware. While WiFi-based indoor localization has been widely studied, BLE has emerged a key technology for contact-tracing in the current pandemic. To accurately estimate distance using BLE on commercial devices, systems today rely on Receiver Signal Strength Indicator(RSSI) which suffers from sampling bias and multipath effects. We propose a new metric: Packet Reception Probability (PRP) that builds on a counter-intuitive idea that we can exploit packet loss to estimate distance. We localize using a Bayesian-PRP formulation that also incorporates an explicit model of the multipath. To make deployment easy, we do not require any hardware, firmware, or driver-level changes to off-the-shelf devices, and require minimal training. PRP can achieve meter level accuracy with just 6 devices with known locations and 12 training locations. We show that fusing PRP with RSSI is beneficial at short distances < 2m. Beyond 2m, fusion is worse than PRP, as RSSI becomes effectively de-correlated with distance. Robust location accuracy at all distances and ease of deployment with PRP can help enable wide range indoor localization solutions using BLE.
View PDF Chat

Pre-trained Neural Recommenders: A Transferable Zero-Shot Framework for Recommendation Systems

2023-01-01
Junting Wang , Adit Krishnan , Hari Sundaram +1 more
Modern neural collaborative filtering techniques are critical to the success of e-commerce, social media, and content-sharing platforms. However, despite technical advances -- for every new application domain, we need to … Modern neural collaborative filtering techniques are critical to the success of e-commerce, social media, and content-sharing platforms. However, despite technical advances -- for every new application domain, we need to train an NCF model from scratch. In contrast, pre-trained vision and language models are routinely applied to diverse applications directly (zero-shot) or with limited fine-tuning. Inspired by the impact of pre-trained models, we explore the possibility of pre-trained recommender models that support building recommender systems in new domains, with minimal or no retraining, without the use of any auxiliary user or item information. Zero-shot recommendation without auxiliary information is challenging because we cannot form associations between users and items across datasets when there are no overlapping users or items. Our fundamental insight is that the statistical characteristics of the user-item interaction matrix are universally available across different domains and datasets. Thus, we use the statistical characteristics of the user-item interaction matrix to identify dataset-independent representations for users and items. We show how to learn universal (i.e., supporting zero-shot adaptation without user or item auxiliary information) representations for nodes and edges from the bipartite user-item interaction graph. We learn representations by exploiting the statistical properties of the interaction data, including user and item marginals, and the size and density distributions of their clusters.

CodeScope: An Execution-based Multilingual Multitask Multidimensional Benchmark for Evaluating LLMs on Code Understanding and Generation

2023-01-01
Weixiang Yan , Haitian Liu , Yunkun Wang +7 more
Large Language Models (LLMs) have demonstrated remarkable performance on coding related tasks, particularly on assisting humans in programming and facilitating programming automation. However, existing benchmarks for evaluating the code understanding … Large Language Models (LLMs) have demonstrated remarkable performance on coding related tasks, particularly on assisting humans in programming and facilitating programming automation. However, existing benchmarks for evaluating the code understanding and generation capacities of LLMs suffer from severe limitations. First, most benchmarks are deficient as they focus on a narrow range of popular programming languages and specific tasks, whereas the real-world software development scenarios show dire need to implement systems with multilingual programming environments to satisfy diverse requirements. Practical programming practices also strongly expect multi-task settings for testing coding capabilities of LLMs comprehensively and robustly. Second, most benchmarks also fail to consider the actual executability and the consistency of execution results of the generated code. To bridge these gaps between existing benchmarks and expectations from practical applications, we introduce CodeScope, an execution-based, multilingual, multi-task, multi-dimensional evaluation benchmark for comprehensively gauging LLM capabilities on coding tasks. CodeScope covers 43 programming languages and 8 coding tasks. It evaluates the coding performance of LLMs from three dimensions (perspectives): difficulty, efficiency, and length. To facilitate execution-based evaluations of code generation, we develop MultiCodeEngine, an automated code execution engine that supports 14 programming languages. Finally, we systematically evaluate and analyze 8 mainstream LLMs on CodeScope tasks and demonstrate the superior breadth and challenges of CodeScope for evaluating LLMs on code understanding and generation tasks compared to other benchmarks. The CodeScope benchmark and datasets are publicly available at https://github.com/WeixiangYAN/CodeScope.
View PDF Chat

Friends with Costs and Benefits: Community Formation with Myopic, Boundedly-Rational Actors

2023-01-01
Naina Balepur , Andy Lee , Hari Sundaram
In this paper we address how complex social communities emerge from local decisions by individuals with limited attention and knowledge. This problem is critical; if we understand community formation mechanisms, … In this paper we address how complex social communities emerge from local decisions by individuals with limited attention and knowledge. This problem is critical; if we understand community formation mechanisms, it may be possible to intervene to improve social welfare. We propose an interpretable, novel model for attributed community formation driven by resource-bounded individuals' strategic, selfish behavior. In our stylized model, attributed individuals act strategically in two dimensions: attribute and network structure. Agents are endowed with limited attention, and communication costs limit the number of active connections. In each time step, each agent proposes a new friendship. Agents then accept proposals, decline proposals, or remove friends, consistent with their strategy to maximize payoff. We identify criteria (number of stable triads) for convergence to some community structure and prove that our community formation model converges to a stable network. Ablations justify the ecological validity of our model and show that each aspect of the model is essential. Our empirical results on a physical world microfinance community demonstrate excellent model fits compared to baseline models.

What should I Ask: A Knowledge-driven Approach for Follow-up Questions Generation in Conversational Surveys

2022-01-01
Yubin Ge , Ziang Xiao , Jana Diesner +3 more
Generating follow-up questions on the fly could significantly improve conversational survey quality and user experiences by enabling a more dynamic and personalized survey structure. In this paper, we proposed a … Generating follow-up questions on the fly could significantly improve conversational survey quality and user experiences by enabling a more dynamic and personalized survey structure. In this paper, we proposed a novel task for knowledge-driven follow-up question generation in conversational surveys. We constructed a new human-annotated dataset of human-written follow-up questions with dialogue history and labeled knowledge in the context of conversational surveys. Along with the dataset, we designed and validated a set of reference-free Gricean-inspired evaluation metrics to systematically evaluate the quality of generated follow-up questions. We then propose a two-staged knowledge-driven model for the task, which generates informative and coherent follow-up questions by using knowledge to steer the generation process. The experiments demonstrate that compared to GPT-based baseline models, our two-staged model generates more informative, coherent, and clear follow-up questions.

Audience-Centric Natural Language Generation via Style Infusion

2022-01-01
Samraj Moorjani , Adit Krishnan , Hari Sundaram +2 more
Adopting contextually appropriate, audience-tailored linguistic styles is critical to the success of user-centric language generation systems (e.g., chatbots, computer-aided writing, dialog systems). While existing approaches demonstrate text style transfer (TST) … Adopting contextually appropriate, audience-tailored linguistic styles is critical to the success of user-centric language generation systems (e.g., chatbots, computer-aided writing, dialog systems). While existing approaches demonstrate text style transfer (TST) with large volumes of parallel or non-parallel data, we argue that grounding style on audience-independent external factors is innately limiting for two reasons. First, it is difficult to collect large volumes of audience-specific stylistic data. Second, some stylistic objectives (e.g., persuasiveness, memorability, empathy) are hard to define without audience feedback. In this paper, we propose the novel task of style infusion - infusing the stylistic preferences of audiences in pretrained language generation models. Since humans are better at pairwise comparisons than direct scoring - i.e., is Sample-A more persuasive/polite/empathic than Sample-B - we leverage limited pairwise human judgments to bootstrap a style analysis model and augment our seed set of judgments. We then infuse the learned textual style in a GPT-2 based text generator while balancing fluency and style adoption. With quantitative and qualitative assessments, we show that our infusion approach can generate compelling stylized examples with generic text prompts. We make the anonymized code and data accessible.
View PDF Chat

Beyond Localized Graph Neural Networks: An Attributed Motif Regularization Framework

2020-11-01
Aravind Sankar , Junting Wang , Adit Krishnan +1 more
We present InfoMotif, a new semi-supervised, motif-regularized, learning framework over graphs. We overcome two key limitations of message passing in popular graph neural networks (GNNs): localization (a k-layer GNN cannot … We present InfoMotif, a new semi-supervised, motif-regularized, learning framework over graphs. We overcome two key limitations of message passing in popular graph neural networks (GNNs): localization (a k-layer GNN cannot utilize features outside the k-hop neighborhood of the labeled training nodes) and over-smoothed (structurally indistinguishable) representations. We propose the concept of attributed structural roles of nodes based on their occurrence in different network motifs, independent of network proximity. Two nodes share attributed structural roles if they participate in topologically similar motif instances over co-varying sets of attributes. Further, InfoMotif achieves architecture independence by regularizing the node representations of arbitrary GNNs via mutual information maximization. Our training curriculum dynamically prioritizes multiple motifs in the learning process without relying on distributional assumptions in the underlying graph or the learning task. We integrate three state-of-the-art GNNs in our framework, to show significant gains (3-10% accuracy) across six diverse, real-world datasets. We see stronger gains for nodes with sparse training labels and diverse attributes in local neighborhood structures.
View PDF Chat

Transfer Learning via Contextual Invariants for One-to-Many Cross-Domain Recommendation

2020-07-25
Adit Krishnan , Mahashweta Das , Mangesh Bendre +2 more
The rapid proliferation of new users and items on the social web has aggravated the gray-sheep user/long-tail item challenge in recommender systems. Historically, cross-domain co-clustering methods have successfully leveraged shared … The rapid proliferation of new users and items on the social web has aggravated the gray-sheep user/long-tail item challenge in recommender systems. Historically, cross-domain co-clustering methods have successfully leveraged shared users and items across dense and sparse domains to improve inference quality. However, they rely on shared rating data and cannot scale to multiple sparse target domains (i.e., the one-to-many transfer setting). This, combined with the increasing adoption of neural recommender architectures, motivates us to develop scalable neural layer-transfer approaches for cross-domain learning. Our key intuition is to guide neural collaborative filtering with domain-invariant components shared across the dense and sparse domains, improving the user and item representations learned in the sparse domains. We leverage contextual invariances across domains to develop these shared modules, and demonstrate that with user-item interaction context, we can learn-to-learn informative representation spaces even with sparse interaction data. We show the effectiveness and scalability of our approach on two public datasets and a massive transaction dataset from Visa, a global payments technology company (19% Item Recall, 3x faster vs. training separate models for each domain). Our approach is applicable to both implicit and explicit feedback settings.
View PDF Chat

GroupIM

2020-07-25
Aravind Sankar , Yanhong Wu , Yuhang Wu +3 more
We study the problem of making item recommendations to ephemeral groups, which comprise users with limited or no historical activities together. Existing studies target persistent groups with substantial activity history, … We study the problem of making item recommendations to ephemeral groups, which comprise users with limited or no historical activities together. Existing studies target persistent groups with substantial activity history, while ephemeral groups lack historical interactions. To overcome group interaction sparsity, we propose data-driven regularization strategies to exploit both the preference covariance amongst users who are in the same group, as well as the contextual relevance of users' individual preferences to each group. We make two contributions. First, we present a recommender architecture-agnostic framework GroupIM that can integrate arbitrary neural preference encoders and aggregators for ephemeral group recommendation. Second, we regularize the user-group latent space to overcome group interaction sparsity by: maximizing mutual information between representations of groups and group members; and dynamically prioritizing the preferences of highly informative members through contextual preference weighting. Our experimental results on several real-world datasets indicate significant performance improvements (31-62% relative NDCG@20) over state-of-the-art group recommendation techniques.
View PDF Chat

Discovering Strategic Behaviors for Collaborative Content-Production in Social Networks

2020-04-20
Yuxin Xiao , Adit Krishnan , Hari Sundaram
Some social networks provide explicit mechanisms to allocate social rewards such as reputation based on users' actions, while the mechanism is more opaque in other networks. Nonetheless, there are always … Some social networks provide explicit mechanisms to allocate social rewards such as reputation based on users' actions, while the mechanism is more opaque in other networks. Nonetheless, there are always individuals who obtain greater rewards and reputation than their peers. An intuitive yet important question to ask is whether these successful users employ strategic behaviors to become influential. It might appear that the influencers "have gamed the system." However, it remains difficult to conclude the rationality of their actions due to factors like the combinatorial strategy space, inability to determine payoffs, and resource limitations faced by individuals. The challenging nature of this question has drawn attention from both the theory and data mining communities. Therefore, in this paper, we are motivated to investigate if resource-limited individuals discover strategic behaviors associated with high payoffs when producing collaborative/interactive content in social networks. We propose a novel framework of Dynamic Dual Attention Networks (DDAN) which models individuals' content production strategies through a generative process, under the influence of social interactions involved in the process. Extensive experimental results illustrate the model's effectiveness in user behavior modeling. We make three strong empirical findings: (1) Different strategies give rise to different social payoffs; (2) The best performing individuals exhibit stability in their preference over the discovered strategies, which indicates the emergence of strategic behavior; and (3) The stability of a user's preference is correlated with high payoffs.
View PDF Chat

Quantifying Voter Biases in Online Platforms: An Instrumental Variable Approach

2019-11-07
Himel Dev , Karrie Karahalios , Hari Sundaram
In content-based online platforms, use of aggregate user feedback (say, the sum of votes) is commonplace as the standard for measuring content quality. Use of vote aggregates, however, is at … In content-based online platforms, use of aggregate user feedback (say, the sum of votes) is commonplace as the standard for measuring content quality. Use of vote aggregates, however, is at odds with the existing empirical literature, which suggests that voters are susceptible to different biases-reputation (e.g., of the poster), social influence (e.g., votes thus far), and position (e.g., answer position). Our goal is to quantify, in an observational setting, the degree of these biases in online platforms. Specifically, what are the causal effects of different impression signals-such as the reputation of the contributing user, aggregate vote thus far, and position of content-on a participant's vote on content? We adopt an instrumental variable (IV) framework to answer this question. We identify a set of candidate instruments, carefully analyze their validity, and then use the valid instruments to reveal the effects of the impression signals on votes. Our empirical study using log data from Stack Exchange websites shows that the bias estimates from our IV approach differ from the bias estimates from the ordinary least squares (OLS) method. In particular, OLS underestimates reputation bias (1.6-2.2x for gold badges) and position bias (up to 1.9x for the initial position) and overestimates social influence bias (1.8-2.3x for initial votes). The implications of our work include: redesigning user interface to avoid voter biases; making changes to platforms' policy to mitigate voter biases; detecting other forms of biases in online platforms.
View PDF Chat

Quantifying Voter Biases in Online Platforms

2019-11-07
Himel Dev , Karrie Karahalios , Hari Sundaram
In content-based online platforms, use of aggregate user feedback (say, the sum of votes) is commonplace as the "gold standard" for measuring content quality. Use of vote aggregates, however, is … In content-based online platforms, use of aggregate user feedback (say, the sum of votes) is commonplace as the "gold standard" for measuring content quality. Use of vote aggregates, however, is at odds with the existing empirical literature, which suggests that voters are susceptible to different biases-reputation (e.g., of the poster), social influence (e.g., votes thus far), and position (e.g., answer position). Our goal is to quantify, in an observational setting, the degree of these biases in online platforms. Specifically, what are the causal effects of different impression signals-such as the reputation of the contributing user, aggregate vote thus far, and position of content-on a participant's vote on content? We adopt an instrumental variable (IV) framework to answer this question. We identify a set of candidate instruments, carefully analyze their validity, and then use the valid instruments to reveal the effects of the impression signals on votes. Our empirical study using log data from Stack Exchange websites shows that the bias estimates from our IV approach differ from the bias estimates from the ordinary least squares (OLS) method. In particular, OLS underestimates reputation bias (1.6-2.2x for gold badges) and position bias (up to 1.9x for the initial position) and overestimates social influence bias (1.8-2.3x for initial votes). The implications of our work include: redesigning user interface to avoid voter biases; making changes to platforms' policy to mitigate voter biases; detecting other forms of biases in online platforms.
View PDF Chat

Growing Attributed Networks through Local Processes

2019-05-13
Harshay Shah , Suhansanu Kuma , Hari Sundaram
This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and … This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and resource-constrained mechanisms explain the emergence of structural properties found in real-world networks. We make three contributions. First, we propose a simple and accurate model of attributed network growth that jointly explains the emergence of in-degree, local clustering, clustering-degree relationship and attribute mixing patterns. Second, we make use of biased random walks to develop a model that forms edges locally, without recourse to global information. Third, we account for multiple sociological phenomena: bounded rationality; structural constraints; triadic closure; attribute homophily; preferential attachment. Our experiments show that the proposed Attributed Random Walk (ARW) model accurately preserves network structure and attribute mixing patterns of real-world networks; it improves upon the performance of eight well-known models by a significant margin of 2.5-10 × .
View PDF Chat

Discovering Archetypes to Interpret Evolution of Individual Behavior

2019-01-01
Kanika Narang , Austin Chung , Hari Sundaram +1 more
In this paper, we aim to discover archetypical patterns of individual evolution in large social networks. In our work, an archetype comprises of $\textit{progressive stages}$ of distinct behavior. We introduce … In this paper, we aim to discover archetypical patterns of individual evolution in large social networks. In our work, an archetype comprises of $\textit{progressive stages}$ of distinct behavior. We introduce a novel Gaussian Hidden Markov Model (G-HMM) Cluster to identify archetypes of evolutionary patterns. G-HMMs allow for: near limitless behavioral variation; imposing constraints on how individuals can evolve; different evolutionary rates; and are parsimonious. Our experiments with Academic and StackExchange dataset discover insightful archetypes. We identify four archetypes for researchers: $\textit{Steady}$, $\textit{Diverse, Evolving and Diffuse}$. We observe clear differences in the evolution of male and female researchers within the same archetype. Specifically, women and men differ within an archetype (e.g. Diverse) in how they start, how they transition and the time spent in mid-career. We also found that the differences in grant income are better explained by the differences in archetype than by differences in gender. For StackOverflow, discovered archetypes could be labeled as $\textit{Experts, Seekers, Enthusiasts, and Facilitators}$. We have strong quantitative results with competing baselines for activity prediction and perplexity. For future session prediction, the proposed G-HMM cluster model improves by an average of $32\%$ for different Stack Exchanges and $24\%$ for Academic dataset. Our model also exhibits lower perplexity than the baselines.

Mining Hidden Populations through Attributed Search

2019-01-01
Suhansanu Kumar , Heting Gao , Changyu Wang +2 more
Researchers often query online social platforms through their application programming interfaces (API) to find target populations such as people with mental illness~\cite{De-Choudhury2017} and jazz musicians~\cite{heckathorn2001finding}. Entities of such target population … Researchers often query online social platforms through their application programming interfaces (API) to find target populations such as people with mental illness~\cite{De-Choudhury2017} and jazz musicians~\cite{heckathorn2001finding}. Entities of such target population satisfy a property that is typically identified using an oracle (human or a pre-trained classifier). When the property of the target entities is not directly queryable via the API, we refer to the property as `hidden' and the population as a hidden population. Finding individuals who belong to these populations on social networks is hard because they are non-queryable, and the sampler has to explore from a combinatorial query space within a finite budget limit. By exploiting the correlation between queryable attributes and the population of interest and by hierarchically ordering the query space, we propose a Decision tree-based Thompson sampler (\texttt{DT-TMP}) that efficiently discovers the right combination of attributes to query. Our proposed sampler outperforms the state-of-the-art samplers in online experiments, for example by 54\% on Twitter. When the number of matching entities to a query is known in offline experiments, \texttt{DT-TMP} performs exceedingly well by a factor of 0.9-1.5$\times$ over the baseline samplers. In the future, we wish to explore the option of finding hidden populations by formulating more complex queries.

Report of 2017 NSF Workshop on Multimedia Challenges, Opportunities and Research Roadmaps

2019-01-01
Shih‐Fu Chang , Alexander G. Hauptmann , Louis‐Philippe Morency +7 more
With the transformative technologies and the rapidly changing global R&amp;D landscape, the multimedia and multimodal community is now faced with many new opportunities and uncertainties. With the open source dissemination … With the transformative technologies and the rapidly changing global R&amp;D landscape, the multimedia and multimodal community is now faced with many new opportunities and uncertainties. With the open source dissemination platform and pervasive computing resources, new research results are being discovered at an unprecedented pace. In addition, the rapid exchange and influence of ideas across traditional discipline boundaries have made the emphasis on multimedia multimodal research even more important than before. To seize these opportunities and respond to the challenges, we have organized a workshop to specifically address and brainstorm the challenges, opportunities, and research roadmaps for MM research. The two-day workshop, held on March 30 and 31, 2017 in Washington DC, was sponsored by the Information and Intelligent Systems Division of the National Science Foundation of the United States. Twenty-three (23) invited participants were asked to review and identify research areas in the MM field that are most important over the next 10-15 year timeframe. Important topics were selected through discussion and consensus, and then discussed in depth in breakout groups. Breakout groups reported initial discussion results to the whole group, who continued with further extensive deliberation. For each identified topic, a summary was produced after the workshop to describe the main findings, including the state of the art, challenges, and research roadmaps planned for the next 5, 10, and 15 years in the identified area.

An Induced Multi-Relational Framework for Answer Selection in Community Question Answer Platforms

2019-01-01
Kanika Narang , Chaoqi Yang , Adit Krishnan +3 more
This paper addresses the question of identifying the best candidate answer to a question on Community Question Answer (CQA) forums. The problem is important because Individuals often visit CQA forums … This paper addresses the question of identifying the best candidate answer to a question on Community Question Answer (CQA) forums. The problem is important because Individuals often visit CQA forums to seek answers to nuanced questions. We develop a novel induced relational graph convolutional network (IR-GCN) framework to address the question. We make three contributions. First, we introduce a modular framework that separates the construction of the graph with the label selection mechanism. We use equivalence relations to induce a graph comprising cliques and identify two label assignment mechanisms---label contrast, label sharing. Then, we show how to encode these assignment mechanisms in GCNs. Second, we show that encoding contrast creates discriminative magnification---enhancing the separation between nodes in the embedding space. Third, we show a surprising result---boosting techniques improve learning over familiar stacking, fusion, or aggregation approaches for neural architectures. We show strong results over the state-of-the-art neural baselines in extensive experiments on 50 StackExchange communities.

The Size Conundrum: Why Online Knowledge Markets Can Fail at Scale

2018-04-23
Himel Dev , Chase Geigle , Qingtao Hu +2 more
In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge … In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge market framing allows site operators to reason about market failures, and to design policies to prevent them. Our goal is to provide insights on large-scale knowledge market failures through an interpretable model. We explore a set of interpretable economic production models on a large empirical dataset to analyze the dynamics of content generation in knowledge markets. Amongst these, the Cobb-Douglas model best explains empirical data and provides an intuitive explanation for content generation through the concepts of elasticity and diminishing returns. Content generation depends on user participation and also on how specific types of content (e.g. answers) depends on other types (e.g. questions). We show that these factors of content generation have constant elasticity and a percentage increase in any of the inputs leads to a constant percentage increase in the output. Furthermore, markets exhibit diminishing returns-the marginal output decreases as the input is incrementally increased. Knowledge markets also vary on their returns to scale-the increase in output resulting from a proportionate increase in all inputs. Importantly, many knowledge markets exhibit diseconomies of scale-measures of market health (e.g., the percentage of questions with an accepted answer) decrease as a function of the number of participants. The implications of our work are two-fold: site operators ought to design incentives as a function of system size (number of participants); the market lens should shed insight into complex dependencies amongst different content types and participant actions in general social networks.
View PDF Chat

Resource Allocation Game on Social Networks: Best Response Dynamics and Convergence

2018-01-01
Wei-Chun Lee , Vasilis Livanos , Ruta Mehta +1 more
The decisions that human beings make to allocate time has significant bearing on economic output and to the sustenance of social networks. The time allocation problem motivates our formal analysis … The decisions that human beings make to allocate time has significant bearing on economic output and to the sustenance of social networks. The time allocation problem motivates our formal analysis of the resource allocation game, where agents on a social network, who have asymmetric, private interaction preferences, make decisions on how to allocate time, a bounded endowment, over their neighbors. Unlike the well-known opinion formation game on a social network, our game appears not to be a potential game, and the Best-Response dynamics is non-differentiable making the analysis of Best-Response dynamics non-trivial. In our game, we consider two types of player behavior, namely optimistic or pessimistic, based on how they use their time endowment over their neighbors. To analyze Best-Response dynamics, we circumvent the problem of the game not being a potential game, through the lens of a novel two-level potential function approach. We show that the Best-Response dynamics converges point-wise to a Nash Equilibrium when players are all: optimistic; pessimistic; a mix of both types. Finally, we show that the Nash Equilibrium set is non-convex but connected, and Price of Anarchy is unbounded while Price of Stability is one. Extensive simulations over a stylized grid reveals that the distribution of quality of the convergence points is unimodal-we conjecture that presence of unimodality is tied to the connectedness of Nash Equilibrium.

The Size Conundrum

2018-01-01
Himel Dev , Chase Geigle , Qingtao Hu +2 more
In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge … In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge market framing allows site operators to reason about market failures, and to design policies to prevent them. Our goal is to provide insights on large-scale knowledge market failures through an interpretable model. We explore a set of interpretable economic production models on a large empirical dataset to analyze the dynamics of content generation in knowledge markets. Amongst these, the Cobb-Douglas model best explains empirical data and provides an intuitive explanation for content generation through concepts of elasticity and diminishing returns. Content generation depends on user participation and also on how specific types of content (e.g. answers) depends on other types (e.g. questions). We show that these factors of content generation have constant elasticity---a percentage increase in any of the inputs leads to a constant percentage increase in the output. Furthermore, markets exhibit diminishing returns---the marginal output decreases as the input is incrementally increased. Knowledge markets also vary on their returns to scale---the increase in output resulting from a proportionate increase in all inputs. Importantly, many knowledge markets exhibit diseconomies of scale---measures of market health (e.g., the percentage of questions with an accepted answer) decrease as a function of number of participants. The implications of our work are two-fold: site operators ought to design incentives as a function of system size (number of participants); the market lens should shed insight into complex dependencies amongst different content types and participant actions in general social networks.
View PDF Chat

Modeling Network Growth under Resource Constraints.

2017-12-29
Harshay Shah , Suhansanu Kumar , Hari Sundaram
We propose a resource-constrained network growth model that explains the emergence of key structural properties of real-world directed networks: heavy-tailed indegree distribution, high local clustering and degree-clustering relationship. In real-world … We propose a resource-constrained network growth model that explains the emergence of key structural properties of real-world directed networks: heavy-tailed indegree distribution, high local clustering and degree-clustering relationship. In real-world networks, individuals form edges under constraints of limited network access and partial information. However, well-known growth models that preserve multiple structural properties do not incorporate these resource constraints. Conversely, existing resource-constrained models do not jointly preserve multiple structural properties of real-world networks. We propose a random walk growth model that explains how real-world network properties can jointly arise from edge formation under resource constraints. In our model, each node that joins the network selects a seed node from which it initiates a random walk. At each step of the walk, the new node either jumps back to the seed node or chooses an outgoing or incoming edge to visit another node. It links to each visited node with some probability and stops after forming a few edges. Our experimental results against four well-known growth models indicate improvement in accurately preserving structural properties of five citation networks. Our model also preserves two structural properties that most growth models cannot: skewed local clustering distribution and bivariate indegree-clustering relationship.

Growing Attributed Networks through Local Processes.

2017-12-29
Harshay Shah , Suhansanu Kumar , Hari Sundaram
This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and … This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and resource-constrained mechanisms explain the emergence of rich structural properties found in real-world networks. We make three contributions. First, we propose a parsimonious and accurate model of attributed network growth that jointly explains the emergence of in-degree distributions, local clustering, clustering-degree relationship and attribute mixing patterns. Second, our model is based on biased random walks and uses local processes to form edges without recourse to global network information. Third, we account for multiple sociological phenomena: bounded rationality, structural constraints, triadic closure, attribute homophily, and preferential attachment. Our experiments indicate that the proposed Attributed Random Walk (ARW) model accurately preserves network structure and attribute mixing patterns of six real-world networks; it improves upon the performance of eight state-of-the-art models by a statistically significant margin of 2.5-10x.

The Size Conundrum: Why Online Knowledge Markets Can Fail at Scale

2017-12-01
Himel Dev , Chase Geigle , Qingtao Hu +2 more
In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge … In this paper, we interpret the community question answering websites on the StackExchange platform as knowledge markets, and analyze how and why these markets can fail at scale. A knowledge market framing allows site operators to reason about market failures, and to design policies to prevent them. Our goal is to provide insights on large-scale knowledge market failures through an interpretable model. We explore a set of interpretable economic production models on a large empirical dataset to analyze the dynamics of content generation in knowledge markets. Amongst these, the Cobb-Douglas model best explains empirical data and provides an intuitive explanation for content generation through concepts of elasticity and diminishing returns. Content generation depends on user participation and also on how specific types of content (e.g. answers) depends on other types (e.g. questions). We show that these factors of content generation have constant elasticity---a percentage increase in any of the inputs leads to a constant percentage increase in the output. Furthermore, markets exhibit diminishing returns---the marginal output decreases as the input is incrementally increased. Knowledge markets also vary on their returns to scale---the increase in output resulting from a proportionate increase in all inputs. Importantly, many knowledge markets exhibit diseconomies of scale---measures of market health (e.g., the percentage of questions with an accepted answer) decrease as a function of number of participants. The implications of our work are two-fold: site operators ought to design incentives as a function of system size (number of participants); the market lens should shed insight into complex dependencies amongst different content types and participant actions in general social networks.
View PDF Chat

Improving Latent User Models in Online Social Media.

2017-11-30
Adit Krishnan , Ashish Sharma , Hari Sundaram
Modern social platforms are characterized by the presence of rich user-behavior data associated with the publication, sharing and consumption of textual content. Users interact with content and with each other … Modern social platforms are characterized by the presence of rich user-behavior data associated with the publication, sharing and consumption of textual content. Users interact with content and with each other in a complex and dynamic social environment while simultaneously evolving over time. In order to effectively characterize users and predict their future behavior in such a setting, it is necessary to overcome several challenges. Content heterogeneity and temporal inconsistency of behavior data result in severe sparsity at the user level. In this paper, we propose a novel mutual-enhancement framework to simultaneously partition and learn latent activity profiles of users. We propose a flexible user partitioning approach to effectively discover rare behaviors and tackle user-level sparsity. We extensively evaluate the proposed framework on massive datasets from real-world platforms including Q&A networks and interactive online courses (MOOCs). Our results indicate significant gains over state-of-the-art behavior models ( 15% avg ) in a varied range of tasks and our gains are further magnified for users with limited interaction data. The proposed algorithms are amenable to parallelization, scale linearly in the size of datasets, and provide flexibility to model diverse facets of user behavior.

Finding by Counting: A Probabilistic Packet Count Model for Indoor Localization in BLE Environments

2017-10-20
Subham De , Shreyans Chowdhary , Aniket Shirke +3 more
We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor … We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor localization to play an important role in indoor public spaces in the future. We model the probability of reception of a packet as a generalized quadratic function of distance, beacon power and advertising frequency. Then, we use a Bayesian formulation to determine the coefficients of the packet loss model using empirical observations from our testbed. We develop a new sequential Monte-Carlo algorithm that uses our packet count model. The algorithm is general enough to accommodate different spatial configurations. We have good indoor localization experiments: our approach has an average error of ~ 1.2m, 53% lower than the baseline range-free Monte-Carlo localization algorithm.
View PDF Chat

Finding by Counting

2017-10-06
Subham De , Shreyans Chowdhary , Aniket Shirke +3 more
We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor … We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor localization to play an important role in indoor public spaces in the future. We model the probability of reception of a packet as a generalized quadratic function of distance, beacon power and advertising frequency. Then, we use a Bayesian formulation to determine the coefficients of the packet loss model using empirical observations from our testbed. We develop a new sequential Monte-Carlo algorithm that uses our packet count model. The algorithm is general enough to accommodate different spatial configurations. We have good indoor localization experiments: our approach has an average error of ~ 1.2m, 53% lower than the baseline range-free Monte-Carlo localization algorithm.
View PDF Chat

Finding by Counting: A Probabilistic Packet Count Model for Indoor Localization in BLE Environments

2017-08-27
Subham De , Shreyans Chowdhary , Aniket Shirke +3 more
We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor … We propose a probabilistic packet reception model for Bluetooth Low Energy (BLE) packets in indoor spaces and we validate the model by using it for indoor localization. We expect indoor localization to play an important role in indoor public spaces in the future. We model the probability of reception of a packet as a generalized quadratic function of distance, beacon power and advertising frequency. Then, we use a Bayesian formulation to determine the coefficients of the packet loss model using empirical observations from our testbed. We develop a new sequential Monte-Carlo algorithm that uses our packet count model. The algorithm is general enough to accommodate different spatial configurations. We have good indoor localization experiments: our approach has an average error of ~1.2m, 53% lower than the baseline range-free Monte-Carlo localization algorithm.
View PDF Chat

Growing Attributed Networks through Local Processes

2017-01-01
Harshay Shah , Suhansanu Kumar , Hari Sundaram
This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and … This paper proposes an attributed network growth model. Despite the knowledge that individuals use limited resources to form connections to similar others, we lack an understanding of how local and resource-constrained mechanisms explain the emergence of rich structural properties found in real-world networks. We make three contributions. First, we propose a parsimonious and accurate model of attributed network growth that jointly explains the emergence of in-degree distributions, local clustering, clustering-degree relationship and attribute mixing patterns. Second, our model is based on biased random walks and uses local processes to form edges without recourse to global network information. Third, we account for multiple sociological phenomena: bounded rationality, structural constraints, triadic closure, attribute homophily, and preferential attachment. Our experiments indicate that the proposed Attributed Random Walk (ARW) model accurately preserves network structure and attribute mixing patterns of six real-world networks; it improves upon the performance of eight state-of-the-art models by a statistically significant margin of 2.5-10x.
View PDF Chat

Improving Latent User Models in Online Social Media

2017-01-01
Adit Krishnan , Ashish Sharma , Hari Sundaram
Modern social platforms are characterized by the presence of rich user-behavior data associated with the publication, sharing and consumption of textual content. Users interact with content and with each other … Modern social platforms are characterized by the presence of rich user-behavior data associated with the publication, sharing and consumption of textual content. Users interact with content and with each other in a complex and dynamic social environment while simultaneously evolving over time. In order to effectively characterize users and predict their future behavior in such a setting, it is necessary to overcome several challenges. Content heterogeneity and temporal inconsistency of behavior data result in severe sparsity at the user level. In this paper, we propose a novel mutual-enhancement framework to simultaneously partition and learn latent activity profiles of users. We propose a flexible user partitioning approach to effectively discover rare behaviors and tackle user-level sparsity. We extensively evaluate the proposed framework on massive datasets from real-world platforms including Q&A networks and interactive online courses (MOOCs). Our results indicate significant gains over state-of-the-art behavior models ( 15% avg ) in a varied range of tasks and our gains are further magnified for users with limited interaction data. The proposed algorithms are amenable to parallelization, scale linearly in the size of datasets, and provide flexibility to model diverse facets of user behavior.
View PDF Chat

Influencing Busy People in a Social Network

2016-10-06
Kaushik Sarkar , Hari Sundaram
We identify influential early adopters in a social network, where individuals are resource constrained, to maximize the spread of multiple, costly behaviors. A solution to this problem is especially important … We identify influential early adopters in a social network, where individuals are resource constrained, to maximize the spread of multiple, costly behaviors. A solution to this problem is especially important for viral marketing. The problem of maximizing influence in a social network is challenging since it is computationally intractable. We make three contributions. First, propose a new model of collective behavior that incorporates individual intent, knowledge of neighbors actions and resource constraints. Second, we show that the multiple behavior influence maximization is NP-hard. Furthermore, we show that the problem is submodular, implying the existence of a greedy solution that approximates the optimal solution to within a constant. However, since the greedy algorithm is expensive for large networks, we propose efficient heuristics to identify the influential individuals, including heuristics to assign behaviors to the different early adopters. We test our approach on synthetic and real-world topologies with excellent results. We evaluate the effectiveness under three metrics: unique number of participants, total number of active behaviors and network resource utilization. Our heuristics produce 15-51% increase in expected resource utilization over the naive approach.
View PDF Chat

It's just a matter of perspective(s): Crowd-Powered Consensus Organization of Corpora

2016-01-01
Ayush Jain , Joon Young Seo , Karan Goel +3 more
We study the problem of organizing a collection of objects - images, videos - into clusters, using crowdsourcing. This problem is notoriously hard for computers to do automatically, and even … We study the problem of organizing a collection of objects - images, videos - into clusters, using crowdsourcing. This problem is notoriously hard for computers to do automatically, and even with crowd workers, is challenging to orchestrate: (a) workers may cluster based on different latent hierarchies or perspectives; (b) workers may cluster at different granularities even when clustering using the same perspective; and (c) workers may only see a small portion of the objects when deciding how to cluster them (and therefore have limited understanding of the "big picture"). We develop cost-efficient, accurate algorithms for identifying the consensus organization (i.e., the organizing perspective most workers prefer to employ), and incorporate these algorithms into a cost-effective workflow for organizing a collection of objects, termed ORCHESTRA. We compare our algorithms with other algorithms for clustering, on a variety of real-world datasets, and demonstrate that ORCHESTRA organizes items better and at significantly lower costs.

Task-driven sampling of attributed networks

2016-01-01
Suhansanu Kumar , Hari Sundaram
This paper introduces new techniques for sampling attributed networks to support standard Data Mining tasks. The problem is important for two reasons. First, it is commonplace to perform data mining … This paper introduces new techniques for sampling attributed networks to support standard Data Mining tasks. The problem is important for two reasons. First, it is commonplace to perform data mining tasks such as clustering and classification of network attributes (attributes of the nodes, including social media posts). Furthermore, the extraordinarily large size of real-world networks necessitates that we work with a smaller graph sample. Second, while random sampling will provide an unbiased estimate of content, random access is often unavailable for many networks. Hence, network samplers such as Snowball sampling, Forest Fire, Random Walk, Metropolis-Hastings Random Walk are widely used; however, these attribute-agnostic samplers were designed to capture salient properties of network structure, not node content. The latter is critical for clustering and classification tasks. There are three contributions of this paper. First, we introduce several attribute-aware samplers based on Information Theoretic principles. Second, we prove that these samplers have a bias towards capturing new content, and are equivalent to uniform sampling in the limit. Finally, our experimental results over large real-world datasets and synthetic benchmarks are insightful: attribute-aware samplers outperform both random sampling and baseline attribute-agnostic samplers by a wide margin in clustering and classification tasks.

Leveraging Social Foci for Information Seeking in Social Media

2015-02-09
Suhas Ranganath , Jiliang Tang , Xia Hu +2 more
The rise of social media provides a great opportunity for people to reach out to their social connections to satisfy their information needs. However, generic social media platforms are not … The rise of social media provides a great opportunity for people to reach out to their social connections to satisfy their information needs. However, generic social media platforms are not explicitly designed to assist information seeking of users. In this paper, we propose a novel framework to identify the social connections of a user able to satisfy his information needs. The information need of a social media user is subjective and personal, and we investigate the utility of his social context to identify people able to satisfy it. We present questions users post on Twitter as instances of information seeking activities in social media. We infer soft community memberships of the asker and his social connections by integrating network and content information. Drawing concepts from the social foci theory, we identify answerers whose community memberships in the question domain overlap with that of the asker. Our experiments demonstrate that the framework is effective in identifying answerers to social media questions.
View PDF Chat

Leveraging Social Foci for Information Seeking in Social Media

2015-01-01
Suhas Ranganath , Jiliang Tang , Xia Hu +2 more
The rise of social media provides a great opportunity for people to reach out to their social connections to satisfy their information needs. However, generic social media platforms are not … The rise of social media provides a great opportunity for people to reach out to their social connections to satisfy their information needs. However, generic social media platforms are not explicitly designed to assist information seeking of users. In this paper, we propose a novel framework to identify the social connections of a user able to satisfy his information needs. The information need of a social media user is subjective and personal, and we investigate the utility of his social context to identify people able to satisfy it. We present questions users post on Twitter as instances of information seeking activities in social media. We infer soft community memberships of the asker and his social connections by integrating network and content information. Drawing concepts from the social foci theory, we identify answerers who share communities with the asker w.r.t. the question. Our experiments demonstrate that the framework is effective in identifying answerers to social media questions.
View PDF Chat

Network Cartography: Seeing the Forest and the Trees

2015-01-01
Jia Wang , Kevin Chen–Chuan Chang , Hari Sundaram
Real-world networks are often complex and large with millions of nodes, posing a great challenge for analysts to quickly see the big picture for more productive subsequent analysis. We aim … Real-world networks are often complex and large with millions of nodes, posing a great challenge for analysts to quickly see the big picture for more productive subsequent analysis. We aim at facilitating exploration of node-attributed networks by creating representations with conciseness, expressiveness, interpretability, and multi-resolution views. We develop such a representation as a {\it map} --- among the first to explore principled network cartography for general networks. In parallel with common maps, ours has landmarks, which aggregate nodes homogeneous in their traits and interactions with nodes elsewhere, and roads, which represent the interactions between the landmarks. We capture such homogeneity by the similar roles the nodes played. Next, to concretely model the landmarks, we propose a probabilistic generative model of networks with roles as latent factors. Furthermore, to enable interactive zooming, we formulate novel model-based constrained optimization. Then, we design efficient linear-time algorithms for the optimizations. Experiments using real-world and synthetic networks show that our method produces more expressive maps than existing methods, with up to 10 times improvement in network reconstruction quality. We also show that our method extracts landmarks with more homogeneous nodes, with up to 90\% improvement in the average attribute/link entropy among the nodes over each landmark. Sense-making of a real-world network using a map computed by our method qualitatively verify the effectiveness of our method.

How Do We Find Early Adopters Who Will Guide a Resource Constrained Network Towards a Desired Distribution of Behaviors?

2013-01-01
Kaushik Sarkar , Hari Sundaram
We identify influential early adopters that achieve a target behavior distribution for a resource constrained social network with multiple costly behaviors. This problem is important for applications ranging from collective … We identify influential early adopters that achieve a target behavior distribution for a resource constrained social network with multiple costly behaviors. This problem is important for applications ranging from collective behavior change to corporate viral marketing campaigns. In this paper, we propose a model of diffusion of multiple behaviors when individual participants have resource constraints. Individuals adopt the set of behaviors that maximize their utility subject to available resources. We show that the problem of influence maximization for multiple behaviors is NP-complete. Thus we propose heuristics, which are based on node degree and expected immediate adoption, to select early adopters. We evaluate the effectiveness under three metrics: unique number of participants, total number of active behaviors and network resource utilization. We also propose heuristics to distribute the behaviors amongst the early adopters to achieve a target distribution in the population. We test our approach on synthetic and real-world topologies with excellent results. Our heuristics produce 15-51\% increase in resource utilization over the naïve approach.

Social event detection with interaction graph modeling

2012-10-29
Yanxiang Wang , Hari Sundaram , Lexing Xie
This paper focuses on detecting social, physical-world events from photos posted on social media sites. The problem is important: cheap media capture devices have significantly increased the number of photos … This paper focuses on detecting social, physical-world events from photos posted on social media sites. The problem is important: cheap media capture devices have significantly increased the number of photos shared on these sites. The main contribution of this paper is to incorporate online social interaction features in the detection of physical events. We believe that online social interaction reflect important signals among the participants on the "social affinity" of two photos, thereby helping event detection. We compute social affinity via a random-walk on a social interaction graph to determine similarity between two photos on the graph. We train a support vector machine classifier to combine the social affinity between photos and photo-centric metadata including time, location, tags and description. Incremental clustering is then used to group photos to event clusters. We have very good results on two large scale real-world datasets: Upcoming and MediaEval. We show an improvement between 0.06-0.10 in F1 on these datasets.
View PDF Chat

Social Event Detection with Interaction Graph Modeling

2012-01-01
Yanxiang Wang , Hari Sundaram , Lexing Xie
This paper focuses on detecting social, physical-world events from photos posted on social media sites. The problem is important: cheap media capture devices have significantly increased the number of photos … This paper focuses on detecting social, physical-world events from photos posted on social media sites. The problem is important: cheap media capture devices have significantly increased the number of photos shared on these sites. The main contribution of this paper is to incorporate online social interaction features in the detection of physical events. We believe that online social interaction reflect important signals among the participants on the "social affinity" of two photos, thereby helping event detection. We compute social affinity via a random-walk on a social interaction graph to determine similarity between two photos on the graph. We train a support vector machine classifier to combine the social affinity between photos and photo-centric metadata including time, location, tags and description. Incremental clustering is then used to group photos to event clusters. We have very good results on two large scale real-world datasets: Upcoming and MediaEval. We show an improvement between 0.06-0.10 in F1 on these datasets.
View PDF Chat
Coauthor Papers Together
Adit Krishnan 10
Junting Wang 5
Karrie Karahalios 5
Himel Dev 5
Suhansanu Kumar 5
Samraj Moorjani 4
Aravind Sankar 4
Subham De 4
Robin Kravets 4
Lexing Xie 3
Shreyans Chowdhary 3
Qingtao Hu 3
Yat Long Lo 3
Chase Geigle 3
Jiahui Zheng 3
Aniket Shirke 3
Harshay Shah 3
Yunzhe Li 2
Ewa Masłowska 2
Ziang Xiao 2
Weixiang Yan 2
Kanika Narang 2
Yunzhe Li 2
Praneet Rathi 2
Kevin Chen–Chuan Chang 2
Jiliang Tang 2
Suhas Ranganath 2
Naina Balepur 2
Kaushik Sarkar 2
Huan Liu 2
Xia Hu 2
Hao Yang 2
Yanxiang Wang 2
Zhining Liu 1
Deepak Vasisht 1
Tiffany Wenting Li 1
Aisling Kelliher 1
Wei-Chun Lee 1
Michelle X. Zhou 1
Andrew Kuznetsov 1
Aditya Parameswaran 1
Ali Ebrahimpour Boroojeny 1
Chaoqi Yang 1
Joyce Chai 1
Balakrishnan Prabhakaran 1
Shrikanth Narayanan 1
A.W.M. Smeulders 1
Shuiguang Deng 1
Jana Diesner 1
Vasilis Livanos 1