Community Detection in Signed Social Networks Using Multiobjective Genetic Algorithm Nancy Girdhar School of Computer and Systems Sciences, Jawaharlal Nehru University, New Delhi, 110067, India. E-mail: nancy.gr1991@gmail.com K. K. Bharadwaj School of Computer and Systems Sciences, Jawaharlal Nehru University, New Delhi, 110067, India. E-mail: kbharadwaj@gmail.com Clustering of like-minded users is basically the goal of community detection (CD) in social networks and many researchers have proposed different algorithms for the same. In signed social networks (SSNs) where type of link is also considered besides the links itself, CD aims to par- tition the network in such a way to have less positive inter-connections and less negative intra-connections among communities. So, approaches used for CD in unsigned networks do not perform well when directly applied on signed networks. Most of the CD algorithms are based on single objective optimization criteria of opti- mizing modularity which focuses only on link density without considering the type of links existing in the net- work. In this work, a multiobjective approach for CD in SSNs is proposed considering both the link density as well as the sign of links. Precisely we are developing a method using modularity, frustration and social balance factor as multiple objectives to be optimized (M-F-SBF model). NSGA-II algorithm is used to maintain elitism and diversity in the solutions. Experiments are performed on both existing benchmarked and real-world datasets show that our approach has led to better solutions, clearly indi- cating the effectiveness of our proposed M-F-SBF model. Introduction People maintain different kinds of relationships and artic- ulate their emotions via different actions like chatting and blocking of users on social websites, friendly and unfriendly nature of gossips at workplace; friendship and bullying in schools or colleges; cliques and disputes at nightspots like clubs; and partnerships and competitions in different organi- zations or corporations. We can easily see these social human relations of love-hate, like-dislike, friends-enemy, trust-distrust in social media sites through different activities of users such as liking or disliking of posts of others, rating blogs on different levels or scale, tagging users as “friends” and “foes” or “fan” and “freak” (terms used in Slashdot web- site). These social relations broadly classified as positive (friendly) relations including “liking,” tagging as “friend” or “fan,” rating “high” on scale and negative (antagonistic) rela- tions (Zhang, Lo, Lim, & Prasetyo, 2013) such as “dislik- ing,” tagging as “foe” or “freak,” rating “low” on scale etc. Many social networking sites explicitly show the mixture of positive and negative interactions among the users by indi- cating the sign on link connecting the users. Some of the popular websites are (a) Tech- news related website Slash- dot, where users can tag other users as “friend” and “foe” or as “fan” and “freak.” Here, the reviewers can comment on the blogs of others depending on their liking or disliking. (b) Trust-network of Epinions is the largest product review website in which users can express their views by rating items of different category like hardware, music and TV shows. Not only items, they can also rate other raters depending on their trust or distrust on that rater. These social networks contain the information about the links as well as the type of links exhibited by the users. As these networks are the extension of social networks, containing information about the sign of links besides the information of links itself, a special term has been coined for these networks called as signed social networks (SSNs; Moshirpour et al., 2013). Exploring community structures from these networks can help us to gain unfathomable knowledge and better under- standing of social ties existing among the users constituting the network (Chen, Chuang, & Chiu, 2014; Verma & Bharadwaj, 2017). CD in unsigned networks involves group- ing of those nodes or vertices which are densely connected within groups and sparsely connected between groups. In Received August 3, 2017; revised September 27, 2018; accepted October 21, 2018 © 2019 ASIS&T • Published online Month 00, 2018 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/asi.24164 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY, 00(0):1–17, 2019