The rise of social media has changed the nature of the fashion industry. Influence is no longer concentrated in the hands of an elite few: social networks have distributed power across a broader set of tastemakers. To understand this new landscape of influence, we created FITNet --- a network of the top 10k influencers of the larger Twitter fashion graph. To construct FITNet, we trained a content-based classifier to identify fashion-relevant Twitter accounts. Leveraging this classifier, we estimated the size of Twitter's fashion subgraph, snowball sampled more than 300k fashion-related accounts based on following relationships, and identified the top 10k influencers in the resulting subgraph. We use FITNet to perform a large-scale analysis of fashion influencers, and demonstrate how the network facilitates discovery, surfacing influencers relevant to specific fashion topics that may be of interest to brands, retailers, and media companies.

This article introduces a novel task-independent sampler for attributed networks. The problem is important because while data mining tasks on network content are common, sampling on internet-scale networks is costly. Link-trace samplers such as Snowball sampling, Forest Fire, Random Walk, and Metropolis–Hastings Random Walk are widely used for sampling from networks. The design of these attribute-agnostic samplers focuses on preserving salient properties of network structure, and are not optimized for tasks on node content. This article has three contributions. First, we propose a task-independent, attribute aware link-trace sampler grounded in Information Theory. Our sampler greedily adds to the sample the node with the most informative (i.e., surprising) neighborhood. The sampler tends to rapidly explore the attribute space, maximally reducing the surprise of unseen nodes. Second, we prove that content sampling is an NP-hard problem. A well-known algorithm best approximates the optimization solution within 1 − 1/ e , but requires full access to the entire graph. Third, we show through empirical counterfactual analysis that in many real-world datasets, network structure does not hinder the performance of surprise based link-trace samplers. Experimental results over 18 real-world datasets reveal: surprise-based samplers are sample efficient and outperform the state-of-the-art attribute-agnostic samplers by a wide margin (e.g., 45% performance improvement in clustering tasks).

This paper proposes a novel algorithm to discover hidden individuals in a social network. The problem is increasingly important for social scientists as the populations (e.g., individuals with mental illness) that they study converse online. Since these populations do not use the category (e.g., mental illness) to self-describe, directly querying with text is non-trivial. To by-pass the limitations of network and query re-writing frameworks, we focus on identifying hidden populations through attributed search. We propose a hierarchical Multi-Arm Bandit (DT-TMP) sampler that uses a decision tree coupled with reinforcement learning to query the combinatorial attributed search space by exploring and expanding along high yielding decision-tree branches. A comprehensive set of experiments over a suite of twelve sampling tasks on three online web platforms, and three offline entity datasets reveals that DT-TMP outperforms all baseline samplers by upto a margin of 54% on Twitter and 48% on RateMDs. An extensive ablation study confirms DT-TMP's superior performance under different sampling scenarios.

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 × .

In this paper we evaluate the effect of noise on community scoring and centrality-based parameters with respect to two different aspects of network analysis: (i) sensitivity, that is how the parameter value changes as edges are removed and (ii) reliability in the context of message spreading, that is how the time taken to broadcast a message changes as edges are removed. Our experiments on synthetic and real-world networks and three different noise models demonstrate that for both the aspects over all networks and all noise models, permanence qualifies as the most effective metric. For the sensitivity experiments closeness centrality is a close second. For the message spreading experiments, closeness and betweenness centrality based initiator selection closely competes with permanence. This is because permanence has a dual characteristic where the cumulative permanence over all vertices is sensitive to noise but the ids of the top-rank vertices, which are used to find seeds during message spreading remain relatively stable under noise.

Detection of non-overlapping and overlapping communities are essentially the same problem. However, current algorithms focus either on finding overlapping or non-overlapping communities. We present a generalized framework that can identify both non-overlapping and overlapping communities, without any prior input about the network or its community distribution. To do so, we introduce a vertex-based metric,GenPerm, that quantifies by how much a vertex belongs to each of its constituent communities. Our community detection algorithm is based on maximizing the GenPerm over all the vertices in the network. We demonstrate, through experiments over synthetic and real-world networks, that GenPerm is more effective than other metrics in evaluating community structure. Further, we show that due to its vertex-centric property, GenPerm can be used to unfold several inferences beyond community detection, such as core-periphery analysis and message spreading. Our algorithm for maximizing GenPerm outperforms six state-of-the-art algorithms in accurately predicting the ground-truth labels. Finally, we discuss the problem of resolution limit in overlapping communities and demonstrate that maximizing GenPerm can mitigate this problem.

The impact and significance of a scientific publication is measured mostly by the number of citations it accumulates over the years. Early prediction of the citation profile of research articles is a significant as well as challenging problem. In this paper, we argue that features gathered from the citation contexts of the research papers can be very relevant for citation prediction. Analyzing a massive dataset of nearly 1.5 million computer science articles and more than 26 million citation contexts, we show that average countX (number of times a paper is cited within the same article) and average citeWords (number of words within the citation context) discriminate between various citation ranges as well as citation categories. We use these features in a stratified learning framework for future citation prediction. Experimental results show that the proposed model significantly outperforms the existing citation prediction models by a margin of 8-10% on an average under various experimental settings. Specifically, the features derived from the citation context help in predicting long-term citation behavior.