ONLINE SOCIAL NETWORK MINING: CURRENT TRENDS AND RESEARCH ISSUES

Abstract

In recent years the increasing tendency of people to use Online Social Networks (OSNs) such as Facebook, Twitter, and LinkedIn, have resulted in making different kinds of interactions and relationships which, in turn, have lead to the generation and availability of a huge amount of valuable data that has never been available before. Such huge valuable data can be used in some new, varied, eye-catching, and useful research areas to researchers. Although this research area has received a great deal of attention in the last few years, yet many problems related to mining OSNs is still in its infancy and needs more techniques to be developed in the future for further improvement. However, since data generated from OSN is vast, noisy, distributed and dynamic, this requires appropriate data mining techniques to analyze such large, complex, and frequently changing social media data. Research is being carried out related to various issues in OSN mining such as, influence propagation, expert finding, recommender systems, link prediction, community detection, opinion mining, mood analysis, prediction of trust and distrust among individuals, etc which is depicted in Fig. 1 below. In this section, we introduce only three representative research issues in mining online social networking sites, namely, influence propagation, community detection and link prediction in OSNs, and give a detailed overview of the related work and current status of these issues.