Detecting Suspicious Profiles Using Text Analysis within Social Media

The exponential advancement in information and communication technology has fostered the emergence of new channels for online discussion and has also reduced distances between people. Unfortunately, malicious people take advantage of this technological achievement in the sense that they use it for illegal purposes. In social media, the users produce several and various formats of suspicious posts (text, image, video...) and exchange them online with other people. The data in most social media sites are stored in text format, so in this work we will focus only on text posts. Text mining is an effective way to add semantics aspect to this communication’s form presenting a significant research challenge. Similarity approach is used in text analysis to detect suspicious posts in social media. The evaluation of our proposed approach is done within real posts.