Evaluations of Similarity Measures on VK for Link Prediction

AbstractRecommender system is one of the most important components for many companies and social networks such as Facebook and YouTube. A recommendation system consists of algorithms which allow to predict and recommend friends or products. This paper studies to facilitate finding like-minded people with same interests in social networks. In our research, we used real data from the most popular social network in Russia, VK (Vkontakte). The study is motivated on the assumption that similarity breeds connection. We evaluate well-known similarity measures in the field on our collected VK datasets and find limited performance results. The result shows that majority of users in VK tend not to add possible users with whom they have common acquaintances. We also propose a topology-based similarity measure to predict future friends. Then, we compare our results with the results of other well-known methods and discuss differences.

[1]  JooYoung Lee,et al.  Introducing ADegree: Anonymisation of Social Networks Through Constraint Programming , 2017, MATES@VLDB.

[2]  Wenliang Du,et al.  Automatic Reputation Computation through Document Analysis: A Social Network Approach , 2011, 2011 International Conference on Advances in Social Networks Analysis and Mining.

[3]  Talal Rahwan,et al.  The Game-Theoretic Interaction Index on Social Networks with Applications to Link Prediction and Community Detection , 2015, IJCAI.

[4]  Junggab Son,et al.  A Hybrid Trust Management Framework for Vehicular Social Networks , 2016, CSoNet.

[5]  Rasheed Hussain,et al.  Evolution of Friendship: a case study of MobiClique , 2017, Conf. Computing Frontiers.

[6]  JooYoung Lee,et al.  Agent Perspective Social Networks: Distributed Second Degree Estimation , 2018, Encyclopedia of Social Network Analysis and Mining. 2nd Ed..

[7]  Kristina Lerman,et al.  Portrait of an Online Shopper: Understanding and Predicting Consumer Behavior , 2015, WSDM.

[8]  JooYoung Lee,et al.  Reputation computation in social networks and its applications , 2014 .

[9]  Robin I. M. Dunbar.  Neocortex size as a constraint on group size in primates , 1992 .

[10]  Tina Eliassi-Rad,et al.  A Probabilistic Model for Using Social Networks in Personalized Item Recommendation , 2015, RecSys.

[11]  Yang Liu,et al.  Who Influenced You? Predicting Retweet via Social Influence Locality , 2015, ACM Trans. Knowl. Discov. Data.

[12]  Le Wu,et al.  Modeling Users' Preferences and Social Links in Social Networking Services: A Joint-Evolving Perspective , 2016, AAAI.

[13]  Joo Young Lee.  Social Network Based Reputation Computation and Document Classification , 2018 .

[14]  Vijay K. Devabhaktuni,et al.  Evaluating Link Prediction Accuracy in Dynamic Networks with Added and Removed Edges , 2016, 2016 IEEE International Conferences on Big Data and Cloud Computing (BDCloud), Social Computing and Networking (SocialCom), Sustainable Computing and Communications (SustainCom) (BDCloud-SocialCom-SustainCom).

[15]  Anna Tigunova,et al.  Location Prediction via Social Contents and Behaviors: Location-Aware Behavioral LDA , 2015, 2015 IEEE International Conference on Data Mining Workshop (ICDMW).

[16]  JooYoung Lee,et al.  Estimating the Degrees of Neighboring Nodes in Online Social Networks , 2014, PRIMA.

[17]  Manuel Mazzara,et al.  Link Prediction Using Top-k Shortest Distances , 2017, BICOD.

[18]  Zhepeng Li,et al.  A Survey of Link Recommendation for Social Networks: Methods, Theoretical Foundations, and Future Research Directions , 2015 .