Detecting malicious short URLs on Twitter

Short URLs have gained immense popularity especially in Online Social Networks, blogs, and messages. Short URLs are used to avoid sharing overly long URLs and save limited text space in messages or tweets. Significant number of URLs shared in the Online Social Networks are either shortened with some URL shortening services or not too long. Despite of its potential benefits from genuine usage, attackers use shortened URLs to hide the malicious URLs, which direct users to malicious pages. Although, URL shortening services use some sort of detection mechanism to protect malicious URLs from being shortened, research has found that they fail to do so effectively. These malicious URLs are found to propagate through OSNs, which fail to stop them effectively as well. In this paper, we develop a machine learning classifier to detect malicious short URLs with visible content features, tweet context, and social features from Online Social Network Twitter