Abstrait

MALLET-Privacy Preserving Influencer Mining in Social Media Networks via Hypergraph

Janani K, Narmatha S

ABSTRACT: Online Social Media Networks (OSNs) provide an online service for building social relations among users to share interests, images, audios and videos. A social network service represents each user’s social links such as likes, comments, favorites and tags which are very useful for mining social influence. The social links indicate certain influence in the community. The existing system suffers from analyzing the generic influence but ignoring the more important topic-level influence. Since the content of interest is essentially topic-specific, the underlying social influence is topic-sensitive. To address these restrictions develop a Novel Topic-Sensitive Influencer Mining (TSIM) framework in social networks which aims to mine topic-specific influential nodes in the networks and find topical influential users and images. The influence estimation is achieved by using hyper graph learning approach in which the vertices represent users and images, and the edges represent multi-type relations include visual-textual content relations among images, and social link relations between users and images. Social influence mining is used in real applications like friend suggestion, photo recommendation, expert identification and social search. The proposed algorithm provides privacy framework for each user in Social Networks like Flickr.