Abhijnan Chakraborty

Detecting overlapping communities in folksonomies

Folksonomies like Delicious and LastFm are modeled as tripartite (user-resource-tag) hypergraphs for studying their network properties. Detecting communities of similar nodes from such networks is a challenging problem. Most existing algorithms for community detection in folksonomies assign unique communities to nodes, whereas in reality, users have multiple topical interests and the same resource is often tagged with semantically different tags. The few attempts to detect overlapping communities work on projections of the hypergraph, which results in significant loss of information contained in the original tripartite structure. We propose the first algorithm to detect overlapping communities in folksonomies using the complete hypergraph structure. Our algorithm converts a hypergraph into its corresponding line-graph, using measures of hyperedge similarity, whereby any community detection algorithm on unipartite graphs can be used to produce overlapping communities in the folksonomy. Through extensive experiments on synthetic as well as real folksonomy data, we demonstrate that the proposed algorithm can detect better community structures as compared to existing state-of-the-art algorithms for folksonomies.

Clustering Hypergraphs for Discovery of Overlapping Communities in Folksonomies

Some of the most popular sites in the Web today are the social tagging systems or folksonomies (e.g. Delicious, Flickr LastFm) where the users share resources and collaboratively annotate those resources with meaningful tags. This helps in the search and the organization of the vast amount of resources. Folksonomies are modelled as tripartite user-resource-tag hypergraphs to study their network properties. Detecting communities of similar nodes from such networks is a challenging and well-studied problem. However, most existing algorithms for community detection in folksonomies assign unique communities to nodes, whereas in reality, nodes are often associated with multiple overlapping communities. Users have multiple topical interests, and the same resource is often tagged with semantically different tags. The few attempts to detect overlapping communities work on projections of the hypergraph, which results in significant loss of the information contained in the original tripartite structure. In this chapter, we present “Overlapping Hypergraph Clustering” algorithm which detects overlapping communities in folksonomies using the complete tripartite hypergraph structure. The algorithm converts a hypergraph into its corresponding weighted line graph, using measures of hyperedge similarity. Then simple nonoverlapping communities are detected from the line graph, which in turn produce overlapping communities in the folksonomy. Through extensive experiments on synthetic as well as real folksonomy data, we demonstrate that the “Overlapping Hypergraph Clustering” algorithm can detect better community structures in folksonomies as compared to the existing state-of-the-art algorithms.

LMPP: A Large Margin Point Process Combining Reinforcement and Competition for Modeling Hashtag Popularity

Predicting the popularity dynamics of Twitter hashtags has a broad spectrum of applications. Existing works have primarily focused on modeling the popularity of individual tweets rather than the underlying hashtags. As a result, they fail to consider several realistic factors contributing to hashtag popularity. In this paper, we propose Large Margin Point Process (LMPP), a probabilistic framework that integrates hashtag-tweet influence and hashtaghashtag competitions, the two factors which play important roles in hashtag propagation. Furthermore, while considering the hashtag competitions, LMPP looks into the variations of popularity rankings of the competing hashtags across time. Extensive experiments on seven real datasets demonstrate that LMPP outperforms existing popularity prediction approaches by a significant margin. Additionally, LMPP can accurately predict the relative rankings of competing hashtags, offering additional advantage over the state-of-the-art baselines.

Tabloids in the Era of Social Media?: Understanding the Production and Consumption of Clickbaits in Twitter

With the growing shift towards news consumption primarily through social media sites like Twitter, most of the traditional as well as new-age media houses are promoting their news stories by tweeting about them. The competition for user attention in such mediums has led many media houses to use catchy sensational form of tweets to attract more users - a process known as clickbaiting. In this work, using an extensive dataset collected from Twitter, we analyze the social sharing patterns of clickbait and non-clickbait tweets to determine the organic reach of such tweets. We also attempt to study the sections of Twitter users who actively engage themselves in following clickbait and non-clickbait tweets. Comparing the advent of clickbaits with the rise of tabloidization of news, we bring out several important insights regarding the news consumers as well as the media organizations promoting news stories on Twitter.

Identifying and Characterizing Sleeping Beauties on YouTube

The generally accepted notion about popularity dynamics of user generated contents (e.g., tweets, videos) is that such contents attain their peak popularity within first few days and then gradually fade into oblivion. However, analyzing more than 350K videos on YouTube, we find that more than 10% of them obtain their peak popularity after at least one year from being uploaded. We term such videos as Sleeping Beauties and observe that these videos engage users more compared to other videos on YouTube. We further observe that sleeping beauties can retain their popularity to a greater extent following their peak popularity compared to other videos. We believe that recognizing such videos will not only benefit the advertisers, but also the designers of recommendation systems who seek to maximize user satisfaction. Through this interactive poster, we bring the presence and characteristics of sleeping beauties in front of the research community.

#FewThingsAboutIdioms: Understanding Idioms and Its Users in the Twitter Online Social Network

To help users find popular topics of discussion, Twitter periodically publishes ‘trending topics’ (trends) which are the most discussed keywords (e.g., hashtags) at a certain point of time. Inspection of the trends over several months reveals that while most of the trends are related to events in the off-line world, such as popular television shows, sports events, or emerging technologies, a significant fraction are not related to any topic / event in the off-line world. Such trends are usually known as idioms, examples being #4WordsBeforeBreakup, #10thingsIHateAboutYou etc. We perform the first systematic measurement study on Twitter idioms. We find that tweets related to a particular idiom normally do not cluster around any particular topic or event. There are a set of users in Twitter who predominantly discuss idioms – common, not-so-popular, but active users who mostly use Twitter as a conversational platform – as opposed to other users who primarily discuss topical contents. The implication of these findings is that within a single online social network, activities of users may have very different semantics; thus, tasks like community detection and recommendation may not be accomplished perfectly using a single universal algorithm. Specifically, we run two (link-based and content-based) algorithms for community detection on the Twitter social network, and show that idiom oriented users get clustered better in one while topical users in the other. Finally, we build a novel service which shows trending idioms and recommends idiom users to follow.

Fair Sharing for Sharing Economy Platforms

Sharing economy platforms, such as Airbnb, Uber or eBay, are an increasingly common way for people to provide their services to earn a living. Yet, the focus in these platforms is either on the satisfaction of the customers of the service, or on boosting successful business transactions. However, recent studies provide a multitude of reasons to worry about the providers in the sharing economy ecosystems. The concerns range from bad working conditions and worker manipulation to discrimination against minorities. This is worsened by the fact that the algorithms used for matching customers and providers, that de facto decide the amount of exposure each provider receives, are proprietary and non-transparent. In this position paper, we propose a novel framework to think about fairness in the matching mechanisms of online sharing economy platforms. Specifically, we focus on various fairness guarantees from the providers’ perspective. Our notion of fairness relies on the idea that, spread over time, all providers should receive the amount of exposure proportional to their relevance or the utility they provide. We postulate that by not requiring every match to be fair, but rather distributing the fairness over time, we can (i) give better guarantees in terms of the overall benefit for the providers and the customers, (ii) make use of implementations from a long line of research concerned with fair division of constrained resources. Overall, our work takes the first step towards rethinking fairness in online sharing economy systems with an additional emphasis on the well-being of providers, and provides insights into parallels with well-established practical implementations in other domains. ACM Reference format: Abhijnan Chakraborty, Asia J. Biega, Aniko Hannak, and Krishna P. Gummadi. 2017. Fair Sharing for Sharing Economy Platforms. In Proceedings of FATREC Workshop on Responsible Recommendation at RecSys 2017, Como, Italy, August 2017 (FATREC 2017), 4 pages. https://doi.org/10.18122/B2BX2S

Level structure of the odd–odd nucleus 54Mn

The energy levels of 54Mn were populated by a 145 MeV 20Ne beam incident on a 51V target nucleus. This, difficult to access, N = 29 nucleus was produced by projectile break-up followed by incomplete fusion of an alpha particle. The de-exciting gamma transitions were detected using a multi-clover array: Indian National Gamma Array (INGA). The level structure of 54Mn has been extended up to E* 5 MeV and Jπ = 15+ with an addition of nine new γ-transitions. The absence of any regular band-like structure in 54Mn indicates the dominance of the single-particle nature of this odd–odd nucleus. The polarization measurements indicate the possibility of a negative-parity level at 1925 keV.

Level structure of the odd-odd nucleus Mn-54

The energy levels of 54Mn were populated by a 145 MeV 20Ne beam incident on a 51V target nucleus. This, difficult to access, N = 29 nucleus was produced by projectile break-up followed by incomplete fusion of an alpha particle. The de-exciting gamma transitions were detected using a multi-clover array: Indian National Gamma Array (INGA). The level structure of 54Mn has been extended up to E* 5 MeV and Jπ = 15+ with an addition of nine new γ-transitions. The absence of any regular band-like structure in 54Mn indicates the dominance of the single-particle nature of this odd–odd nucleus. The polarization measurements indicate the possibility of a negative-parity level at 1925 keV.

On Designing Content Recommender Systems for Online News Media

Due to the enormous amount of information being carried over online systems today, no user can access all such information. Therefore, to help the users, all major online organizations deploy information retrieval (content recommendation, search or ranking) systems to find important information. Current information retrieval systems have to make certain design choices. For example, news recommendation systems need to decide on the quality of recommended news stories, how much emphasis to give to a storys long-term importance over its recency or freshness etc. Similarly, recommendation systems over user generated contents (e.g., in social media like Facebook and Twitter) need to take into account the content posted by heterogeneous user groups. However, such design choices can introduce unintended biases in the contents presented to the users. For example, the recommended contents may have poor quality or less news value, or the news discourse may get hijacked by hyper-active demographic groups. In this thesis, we want to systematically measure the effect of such design choices in the content recommendation systems, and build alternate recommendation systems that mitigate the biases in the recommendation output.