Detecting Suspicious Users in Social Networks Using Text Analysis

Abstract

Twitter is a social networking platform that allows the users to discuss current news, recent trends, facts, and much more. Meanwhile, due to this, popularity also becomes the target for suspicious users to perform spamming activities via their posts or messages. It is true that social media entice as mechanisms to ease the spread of current news, hate, etc. and allow users to discuss their views by posting tweets as their post, but these services create an opportunity for the attacker to do some dishonest activity. Malicious users mostly post tweets having such topics that may attract us to read, based on some news, but in between, the tweet they usually muddled URL that lead users to entirely unrelated websites or suspicious profiles. This paper examines the text for a suspicious user using profile-based extraction of tweet data based on content as well as features-based attributes. This will help in the analysis of text having with some suspicious activity.