Using communities against deception in online social networks
Sajid Yousuf Bhat, Muhammad Abulaish · Computer Fraud & Security · 2014
Online social networking (OSN) sites such as Facebook and Twitter have become highly popular on the Internet with millions of members sharing information and content, and connecting with each other. The connections thus established reflect the real-world relationships between the users of these social networks. These sites are being looked upon as high-potential marketing opportunities by many organisations. OSNs offer many useful properties that reflect real-world social network characteristics, which include small-world behaviour, significant local clustering, the existence of large, strongly connected components and formation of tightly knit groups or communities.1, 2, 3 Malicious activities in online social networks (OSNs) have transformed from simple spamming to highly deceptive forms focused on breaching the privacy of online social network users and ultimately their trust. Traditional content-based and collaborative filtering techniques give only average results. The topological characteristics of legitimate users, including the formation of tightly knit communities, is a more promising approach, but we need to devise efficient techniques for identifying spammers and attackers, explain Sajid Yousuf Bhat and Muhammad Abulaish of Jamia Millia Islamia, New Delhi.