Detecting Influential Users in a Trending Topic Community Using Link Analysis Approach

Abstract

Social media has become a popular platform to post or share personal information, opinion, photos, videos, etc. Detection of influential users is a significant problem in information diffusion or propagation. Previous researches find influential users based on follower or retweet relationships and centrality measurement approaches. In this paper, influential users are detected by network topology that was obtained from communication relationships among users, link analysis approach, and user's profile features. The proposed approach aims to detect trending topic influencers. Firstly, communication relationships namely retweet, mention, and reply between users in a trending topic are extracted and a trending topic graph is constructed. Secondly, influential users are detected using a link analysis approach combined with the ability of users’ profile features. The performance of the system can be proved to compare with influencer detection methods. The experimental result shows that the trending topic influencers can detect using the interaction relationships and user's features.