Followers Are Not Enough: A Multifaceted Approach to Community Detection in Online Social Networks

Abstract

In online social media networks, individuals often have hundreds or even thousands of connections, which link these users not only to friends, associates, and colleagues, but also to news outlets, celebrities, and organizations. In these complex social networks, a ‘community’ as studied in the social network literature, can have very different meaning depending on the property of the network under study. Taking into account the multifaceted nature of these networks, we claim that community detection in online social networks should also be multifaceted in order to capture all of the different and valuable viewpoints of ‘community.’ In this paper we focus on three types of communities beyond follower-based structural communities: activity-based, topic-based, and interaction-based. We analyze a Twitter dataset using three different weightings of the structural network meant to highlight these three community types, and then infer the communities associated with these weightings. We show that interesting insights can be obtained about the complex community structure present in social networks by studying when and how these four community types give rise to similar as well as completely distinct community structure.

Characterising and evaluating dynamic online communities from live microblogging user interactions

“…However, a fundamental challenge is the diversity in the literature for a definition of user community, which makes community detection difficult to evaluate and interpret. For microblogging OSN, researchers often use the same definition as for more traditional social media (Darmon et al 2015;Bakillah et al 2015;Amor et al 2016;Cao et al 2015), or definitions from topic analysis and user profiling (Zhou et al 2012;Akbari and Chua 2017). However, we argue that these adoptions might not be suitable for microblogging without considering the temporal dynamics of the platform due to its particular fast pace and user sparsity.…”

Anatomy of leadership in collective behaviour

“…These measures are often viewed as less subjective inference methods because almost no assumptions need to be made about the structure of the system being observed. As a result, information theory has become a popular tool for inferring leadership from time series [19][20][21][22][23][24][25][26]62 . However, while influence, information flow and causality are all closely related to the notion of leadership, these concepts are inherently different and therefore are not readily interchangeable.…”

Improving Twitter Community Detection through Contextual Sentiment Analysis

“…For works that aimed to identify communities of people that interacted with each other frequently, main relationships involved are the "re-tweet" and "mention" relationships (Amor et al, 2015;Darmon et al, 2014;Bakillah et al, 2015). When user A re-tweets a tweet by user B, user A is essentially re-publishing the said tweet and propagating it to his/her own followers.…”

“…In addition, SLPA can identify overlapping communities. Darmon et al (2014) chose the Order Statistics Local Optimization Method, or OSLOM, (Lancichinetti et al, 2011) for community detection because of its ability to work with weighted and directed graphs, and its ability to identify overlapping communities. Bakillah et al (2015) chose the Fast-Greedy Optimization of Modularity, or FGM, (Clauset et al, 2004) for its ability to handle complex social graphs from Twitter, and the Varied Density-Based Spatial Clustering of Applications with Noise, or VDBSCAN, (Liu et al, 2007) for its ability to obtain spatial clusters at certain points in time.…”