arXiv subjects

Nitesh V. Chawla

Publications and source records attributed to Nitesh V. Chawla.

118 records · Page 7Linked to original sources

arXiv2015

Modeling the Interplay Between Individual Behavior and Network Distributions

It is well-known that many networks follow a power-law degree distribution; however, the factors that influence the formation of their distributions are still unclear. How can one model the connection between individual actions and network distributions? How can one explain the formation of group phenomena and their evolutionary patterns? In this paper, we propose a unified framework, M3D, to model human dynamics in social networks from three perspectives: macro, meso, and micro. At the micro-level, we seek to capture the way in which an individual user decides whether to perform an action. At the meso-level, we study how group behavior develops and evolves over time, based on individual actions. At the macro-level, we try to understand how network distributions such as power-law (or heavy-tailed phenomena) can be explained by group behavior. We provide theoretical analysis for the proposed framework, and discuss the connection of our framework with existing work. The framework offers a new, flexible way to explain the interplay between individual user actions and network distributions, and can benefit many applications. To model heavy-tailed distributions from partially observed individual actions and to predict the formation of group behaviors, we apply M3D to three different genres of networks: Tencent Weibo, Citation, and Flickr. We also use information-burst prediction as a particular application to quantitatively evaluate the predictive power of the proposed framework. Our results on the Weibo indicate that M3D's prediction performance exceeds that of several alternative methods by up to 30\%.

cs.SI↗

arXiv2015

Evaluating Link Prediction Methods

Link prediction is a popular research area with important applications in a variety of disciplines, including biology, social science, security, and medicine. The fundamental requirement of link prediction is the accurate and effective prediction of new links in networks. While there are many different methods proposed for link prediction, we argue that the practical performance potential of these methods is often unknown because of challenges in the evaluation of link prediction, which impact the reliability and reproducibility of results. We describe these challenges, provide theoretical proofs and empirical examples demonstrating how current methods lead to questionable conclusions, show how the fallacy of these conclusions is illuminated by methods we propose, and develop recommendations for consistent, standard, and applicable evaluation metrics. We also recommend the use of precision-recall threshold curves and associated areas in lieu of receiver operating characteristic curves due to complications that arise from extreme imbalance in the link prediction classification problem.

cs.IR↗

arXiv2015

Inferring Unusual Crowd Events From Mobile Phone Call Detail Records

The pervasiveness and availability of mobile phone data offer the opportunity of discovering usable knowledge about crowd behaviors in urban environments. Cities can leverage such knowledge in order to provide better services (e.g., public transport planning, optimized resource allocation) and safer cities. Call Detail Record (CDR) data represents a practical data source to detect and monitor unusual events considering the high level of mobile phone penetration, compared with GPS equipped and open devices. In this paper, we provide a methodology that is able to detect unusual events from CDR data that typically has low accuracy in terms of space and time resolution. Moreover, we introduce a concept of unusual event that involves a large amount of people who expose an unusual mobility behavior. Our careful consideration of the issues that come from coarse-grained CDR data ultimately leads to a completely general framework that can detect unusual crowd events from CDR data effectively and efficiently. Through extensive experiments on real-world CDR data for a large city in Africa, we demonstrate that our method can detect unusual events with 16% higher recall and over 10 times higher precision, compared to state-of-the-art methods. We implement a visual analytics prototype system to help end users analyze detected unusual crowd events to best suit different application scenarios. To the best of our knowledge, this is the first work on the detection of unusual events from CDR data with considerations of its temporal and spatial sparseness and distinction between user unusual activities and daily routines.

cs.SI↗

arXiv2014

Will This Paper Increase Your h-index? Scientific Impact Prediction

Scientific impact plays a central role in the evaluation of the output of scholars, departments, and institutions. A widely used measure of scientific impact is citations, with a growing body of literature focused on predicting the number of citations obtained by any given publication. The effectiveness of such predictions, however, is fundamentally limited by the power-law distribution of citations, whereby publications with few citations are extremely common and publications with many citations are relatively rare. Given this limitation, in this work we instead address a related question asked by many academic researchers in the course of writing a paper, namely: "Will this paper increase my h-index?" Using a real academic dataset with over 1.7 million authors, 2 million papers, and 8 million citation relationships from the premier online academic service ArnetMiner, we formalize a novel scientific impact prediction problem to examine several factors that can drive a paper to increase the primary author's h-index. We find that the researcher's authority on the publication topic and the venue in which the paper is published are crucial factors to the increase of the primary author's h-index, while the topic popularity and the co-authors' h-indices are of surprisingly little relevance. By leveraging relevant factors, we find a greater than 87.5% potential predictability for whether a paper will contribute to an author's h-index within five years. As a further experiment, we generate a self-prediction for this paper, estimating that there is a 76% probability that it will contribute to the h-index of the co-author with the highest current h-index in five years. We conclude that our findings on the quantification of scientific impact can help researchers to expand their influence and more effectively leverage their position of "standing on the shoulders of giants."

cs.SI↗

arXiv2014

Predicting Node Degree Centrality with the Node Prominence Profile

Centrality of a node measures its relative importance within a network. There are a number of applications of centrality, including inferring the influence or success of an individual in a social network, and the resulting social network dynamics. While we can compute the centrality of any node in a given network snapshot, a number of applications are also interested in knowing the potential importance of an individual in the future. However, current centrality is not necessarily an effective predictor of future centrality. While there are different measures of centrality, we focus on degree centrality in this paper. We develop a method that reconciles preferential attachment and triadic closure to capture a node's prominence profile. We show that the proposed node prominence profile method is an effective predictor of degree centrality. Notably, our analysis reveals that individuals in the early stage of evolution display a distinctive and robust signature in degree centrality trend, adequately predicted by their prominence profile. We evaluate our work across four real-world social networks. Our findings have important implications for the applications that require prediction of a node's future degree centrality, as well as the study of social network dynamics.

cs.SI↗

arXiv2014

Microscopic Evolution of Social Networks by Triad Position Profile

Disentangling the mechanisms underlying the social network evolution is one of social science's unsolved puzzles. Preferential attachment is a powerful mechanism explaining social network dynamics, yet not able to explain all scaling-laws in social networks. Recent advances in understanding social network dynamics demonstrate that several scaling-laws in social networks follow as natural consequences of triadic closure. Macroscopic comparisons between them are discussed empirically in many works. However the network evolution drives not only the emergence of macroscopic scaling but also the microscopic behaviors. Here we exploit two fundamental aspects of the network microscopic evolution: the individual influence evolution and the process of link formation. First we develop a novel framework for the microscopic evolution, where the mechanisms of preferential attachment and triadic closure are well balanced. Then on four real-world datasets we apply our approach for two microscopic problems: node's prominence prediction and link prediction, where our method yields significant predictive improvement over baseline solutions. Finally to be rigorous and comprehensive, we further observe that our framework has a stronger generalization capacity across different kinds of social networks for two microscopic prediction problems. We unveil the significant factors with a greater degree of precision than has heretofore been possible, and shed new light on networks evolution.

cs.SI↗

arXiv2014

Predicting Online Video Engagement Using Clickstreams

In the nascent days of e-content delivery, having a superior product was enough to give companies an edge against the competition. With today's fiercely competitive market, one needs to be multiple steps ahead, especially when it comes to understanding consumers. Focusing on a large set of web portals owned and managed by a private communications company, we propose methods by which these sites' clickstream data can be used to provide a deep understanding of their visitors, as well as their interests and preferences. We further expand the use of this data to show that it can be effectively used to predict user engagement to video streams.

cs.LG↗

arXiv2011

Range-limited Centrality Measures in Complex Networks

Here we present a range-limited approach to centrality measures in both non-weighted and weighted directed complex networks. We introduce an efficient method that generates for every node and every edge its betweenness centrality based on shortest paths of lengths not longer than $\ell = 1,...,L$ in case of non-weighted networks, and for weighted networks the corresponding quantities based on minimum weight paths with path weights not larger than $w_{\ell}=\ell Δ$, $\ell=1,2...,L=R/Δ$. These measures provide a systematic description on the positioning importance of a node (edge) with respect to its network neighborhoods 1-step out, 2-steps out, etc. up to including the whole network. We show that range-limited centralities obey universal scaling laws for large non-weighted networks. As the computation of traditional centrality measures is costly, this scaling behavior can be exploited to efficiently estimate centralities of nodes and edges for all ranges, including the traditional ones. The scaling behavior can also be exploited to show that the ranking top-list of nodes (edges) based on their range-limited centralities quickly freezes as function of the range, and hence the diameter-range top-list can be efficiently predicted. We also show how to estimate the typical largest node-to-node distance for a network of $N$ nodes, exploiting the aforementioned scaling behavior. These observations are illustrated on model networks and on a large social network inferred from cell-phone trace logs ($\sim 5.5\times 10^6$ nodes and $\sim 2.7\times 10^7$ edges). Finally, we apply these concepts to efficiently detect the vulnerability backbone of a network (defined as the smallest percolating cluster of the highest betweenness nodes and edges) and illustrate the importance of weight-based centrality measures in weighted networks in detecting such backbones.

physics.soc-ph↗

arXiv2011

Weighted reciprocity in human communication networks

In this paper we define a metric for reciprocity---the degree of balance in a social relationship---appropriate for weighted social networks in order to investigate the distribution of this dyadic feature in a large-scale system built from trace-logs of over a billion cell-phone communication events across millions of actors. We find that dyadic relations in this network are characterized by much larger degrees of imbalance than we would expect if persons kept only those relationships that exhibited close to full reciprocity. We point to two structural features of human communication behavior and relationship formation---the division of contacts into strong and weak ties and the tendency to form relationships with similar others---that either help or hinder the ability of persons to obtain communicative balance in their relationships. We examine the extent to which deviations from reciprocity in the observed network are partially traceable to these characteristics.

cs.SI↗

arXiv2011

Predictors of short-term decay of cell phone contacts in a large scale communication network

Under what conditions is an edge present in a social network at time t likely to decay or persist by some future time t + Delta(t)? Previous research addressing this issue suggests that the network range of the people involved in the edge, the extent to which the edge is embedded in a surrounding structure, and the age of the edge all play a role in edge decay. This paper uses weighted data from a large-scale social network built from cell-phone calls in an 8-week period to determine the importance of edge weight for the decay/persistence process. In particular, we study the relative predictive power of directed weight, embeddedness, newness, and range (measured as outdegree) with respect to edge decay and assess the effectiveness with which a simple decision tree and logistic regression classifier can accurately predict whether an edge that was active in one time period continues to be so in a future time period. We find that directed edge weight, weighted reciprocity and time-dependent measures of edge longevity are highly predictive of whether we classify an edge as persistent or decayed, relative to the other types of factors at the dyad and neighborhood level.

cs.SI↗