Understanding why people join, stay, or leave social groups is a central question in the social sciences, including computational social systems, while modeling these processes is a challenge in complex networks. Yet, the current empirical studies rarely focus on group dynamics for lack of data relating opinions to group membership. In the NetSense data, we find hundreds of face-to-face groups whose members make thousands of changes of memberships and opinions. We also observe two trends: opinion homogeneity grows over time, and individuals holding unpopular opinions frequently change groups. These observations and data provide us with the basis on which we model the underlying dynamics of human behavior. We formally define the utility that members gain from ingroup interactions as a function of the levels of homophily of opinions of group members with opinions of a given individual in this group. We demonstrate that so-defined utility applied to our empirical data increases after each observed change. We then introduce an analytical model and show that it accurately recreates the trends observed in the NetSense data.

We study community detection in criminal networks and address the problem caused by intentionally hidden edges which hinder the performance of community detection. We make use of link prediction to demonstrate how the community structure of a network can be better identified by augmenting it with edges. We demonstrate the value of this method by showing this method delivers us better quality communities for real life drug trafficking networks. We discuss also the limitations of the approach, and importance of community detection for investigating of criminal networks.

How do opinions of individuals on controversial issues such as marijuana and gay marriage and their underlying social network connections evolve over time? Do people alter their network to have more like-minded friends or do they change their own opinions? Does the society eventually develop echo chambers? In this paper, we study dynamically evolving networks and changing user opinions to answer these questions. Our contributions are as follows: (a) Discovering Evolution of Polarization in Networks: We present evidence of growing divide among users based on their opinions who eventually form homophilic groups (b) Studying Opinion and Network Co-Evolution: We present observations of how individuals change opinions and position themselves in dynamically changing networks (c) Forecasting Persistence and Change in Opinions and Network: We propose ONE-M to forecast individual beliefs and persistence or dissolution of social ties. Using a unique real-world network dataset including periodic user surveys, we show that ONE-M performs with high accuracy, while outperforming the baseline approaches. Code related to this paper is available at: https://github.com/anigam/ONE-M and Data related to this paper is available at: http://netsense.nd.edu/ .

Ashwin Bahulkar∗, Boleslaw K. Szymanski∗, Kevin Chan† and Omar Lizardo‡ ∗Department of Computer Science and Network Science and Technology Center Rensselaer Polytechnic Institute, 110 8th Street, Troy NY 12180, USA Email: bahula@rpi.edu, szymab@rpi.edu †US Army Research Laboratory, Adelphi, MD 20783, USA Email: kevin.s.chan.civ@mail.mil ‡Department of Sociology, UC Los Angeles, Los Angeles, CA 90095 Email: olizardo@ucla.soc.edu

In this paper, we study the interaction patterns among university students whose interactions are recorded in an aligned multilayer social network. One layer of this network represents the smartphone communications, including calls and text messages. The other layer represents face-to- face interactions. We analyze this multilayer network to find whether communication and face-to- face interactions are correlated, and what impact the aging, that is the growing over time familiarity of each node with its peers, has on students' interaction patterns. We also investigate to what extent the academic year structure and external events, such as holidays, affect the network and the interactions between the nodes, and how the individual's communication pattern profile varies as a function of the node's degree and intensity of its interactions. The results that we obtained shed a light on how students' interaction patterns are impacted by the structure of a social network, its age, and social profiles of its nodes.

In this paper, we describe a framework that integrates descriptive, predictive, and prescriptive analytics that aids detecting and disrupting a transnational criminal organization (TCO) operating as interdependent contraband smuggling, money, and money laundering networks. This type of TCO will smuggle contraband across the U.S. border, generate revenues from illegal sales within the U.S., and then use the money laundering network to send the money out of the U.S. Law enforcement may have partial information about the underlying social network of the TCO but this may be missing important, intentionally hidden connections between the criminals. The proposed framework predicts the missing links in the social network data and then algorithms are applied to the augmented data to detect the communities of the TCO. Each community serves a different role in the TCO and thus are necessary in modeling the operations of the organization. Once the communities are identified, we prescribe actions that allocate resources to disrupt the TCO operations optimally in terms of law enforcement criteria.

Previous work has shown that selectivity based on opinions and values of attributes is an important tie-formation mechanism in human social networks. Less well-known is how selectivity influences the formation and composition of whole groups in which interactions extend beyond the dyads. To address this question, we use data from the NetSense study consisting of a multi-layer (nomination, communication, co-location) network of university students. We examine how group formation differs from tie-formation in terms of the role of selectivity based on opinions and attributes. In addition, we show how levels of such selectivity varies between groups formed to meet different needs.

Databases on scientific publications are a well-known source for complex network analysis. The present work focuses on tracking evolution of collaboration amongst researchers on leishmaniasis, a neglected disease associated with poverty and very common in Brazil, India and many other countries in Latin America, Asia and Africa. Using SCOPUS and PubMed databases we have identified clusters of publications resulting from research areas and collaboration between countries. Based on the collaboration patterns, areas of research and their evolution over the past 35 years, we combined different methods in order to understand evolution in science. The methods took into consideration descriptive network analysis combined with lexical analysis of publications, and the collaboration patterns represented by links in network structure. The methods used country of the authors’ publications, MeSH terms, and the collaboration patterns in seven five-year period collaboration network and publication networks snapshots as attributes. The results show that network analysis metrics can bring evidences of evolution of collaboration between different research groups within a specific research area and that those areas have subnetworks that influence collaboration structures and focus.

BACKGROUND:We examine the coevolution of three-layer node-aligned network of university students. The first layer is defined by nominations based on perceived prominence collected from repeated surveys during the first four semesters; the second is a behavioral layer representing actual students' interactions based on records of mobile calls and text messages; while the third is a behavioral layer representing potential face-to-face interactions suggested by bluetooth collocations.METHODS:We address four interrelated questions. First, we ask whether the formation or dissolution of a link in one of the layers precedes or succeeds the formation or dissolution of the corresponding link in another layer (temporal dependencies). Second, we explore the causes of observed temporal dependencies between the layers. For those temporal dependencies that are confirmed, we measure the predictive capability of such dependencies. Third, we observe the progress towards nominations and the stages that lead to them. Finally, we examine whether the differences in dissolution rates of symmetric (undirected) versus asymmetric (directed) links co-exist in all layers.RESULTS:We find strong patterns of reciprocal temporal dependencies between the layers. In particular, the creation of an edge in either behavioral layer generally precedes the formation of a corresponding edge in the nomination layer. Conversely, the decay of a link in the nomination layer generally precedes a decline in the intensity of communication and collocation. Finally, nodes connected by asymmetric nomination edges have lower overall communication and collocation volumes and more asymmetric communication flows than the nodes linked by symmetric edges.CONCLUSION:We find that creation and dissolution of cognitively salient contacts have temporal dependencies with communication and collocation behavior.

We study a unique network dataset including periodic surveys and electronic logs of dyadic contacts via smartphones. The participants were a sample of freshmen entering university in the Fall 2011. Their opinions on a variety of political and social issues and lists of activities on campus were regularly recorded at the beginning and end of each semester for the first three years of study. We identify a behavioral network defined by call and text data, and a cognitive network based on friendship nominations in ego-network surveys. Both networks are limited to study participants. Since a wide range of attributes on each node were collected in self-reports, we refer to these networks as attribute-rich networks. We study whether student preferences for certain attributes of friends can predict formation and dissolution of edges in both networks. We introduce a method for computing student preferences for different attributes which we use to predict link formation and dissolution. We then rank these attributes according to their importance for making predictions. We find that personal preferences, in particular political views, and preferences for common activities help predict link formation and dissolution in both the behavioral and cognitive networks.

We examine the dynamics of co-evolution of two coupled social networks. The first is a cognitive network defined by nominations based on perceived prominence collected from repeated surveys of students during their first four semesters of college while the second is built from the behavioral network representing actual interactions between these individuals based on records of their mobile calls and text messages. We address three interrelated questions. First, we ask whether the formation or dissolution of a link in one of the networks precedes or succeeds formation or dissolution of the corresponding link in the other network (temporal dependencies). Second, we explore the causes of observed temporal dependencies between the two networks. For those temporal dependencies that are confirmed, we measure the predictive capacity of such dependencies. Finally, we examine whether there are systematic differences in the dissolution rates of symmetric (undirected) versus asymmetric (directed) edges in both networks. We find strong patterns of reciprocal temporal dependencies between the two networks. In particular, the creation of an edge in the behavioral network generally precedes the formation of a corresponding edge in the cognitive network. Conversely, the decay of a link in the cognitive network generally precedes a decline in the intensity of communication in the behavioral network. Finally, asymmetric edges in the cognitive network have lower overall communication volume and more asymmetric communication flows in the behavioral network.

CommunityChen, Mingming detectionBahulkar, Ashwin isKuzmin, Konstantin anSzymanski, Boleslaw K. important step of network analysis that relies on the correctness of edges. However, incompleteness and inaccuracy of network data collection methods often cause the communities based on the collected datasets to be different from the ground truth. In this paper, we aim to recover or improve the network community structure using scores provided by different link prediction techniques to replace a fraction of low ranking existing links with top ranked predicted links. Experimental results show that applying our approach to different networks can significantly refine community structure. We also show that predictions of edge additions and persistence are confirmed by the future states of evolving social networks. Another important finding is that not every metric performs equally well on all networks. We observe that performance of link prediction ranking is correlated with certain network properties, such as the network size or average node degree.

We study a unique behavioral network data set (based on periodic surveys and on electronic logs of dyadic contact via smartphones) collected at the University of Notre Dame. The participants are a sample of members of the entering class of freshmen in the fall of 2011 whose opinions on a wide variety of political and social issues and activities on campus were regularly recorded — at the beginning and end of each semester — for the first three years of their residence on campus. We create a communication activity network implied by call and text data, and a friendship network based on surveys. Both networks are limited to students participating in the NetSense surveys. We aim at finding student traits and activities on which agreements correlate well with formation and persistence of links while disagreements is highly correlated with non-existence or dissolution of links in the two social networks that we created. Using statistical analysis and machine learning, we observe several traits and activities displaying such correlations, thus being of potential use to predict social network evolution.

Table is one of the most common mechanisms used for presenting structured information on the web. A table presents information on a set of related concepts in a domain. A column typically represents a concept or an attribute of a concept that the column header identifies. A row contains corresponding instances and attribute values. However column headers are usually quite noisy and sometimes even missing. While a human reader can figure out the required domain mappings relatively easily by using domain knowledge and surrounding context, discovering them algorithmically poses challenges. In this paper we present an algorithm that exploits the idea that a table only presents information on connected entities of a domain ontology. The algorithm works in two phases. In the first phase it uses local optimization criteria such as lexical matching, instance matching, and so on to find an initial set of mappings. In the second phase it takes these mappings and constructs all possible connected sub graphs of the ontology that can be formed from these mappings. The largest of these sub graphs that has the highest local mapping score is then selected as the underlying domain mapping of the table. We present experimental results demonstrating the effectiveness of the algorithm.