Development of a Hack Detection System for Twitter based on Crowd Reaction Analysis

Twitter, a popular microblogging platform, is often target of hackers who take over accounts for sending spam. This triggers a change not only in the accounts behavior itself, but also often in the network connected to it. People that follow the hacked account often might be able to notice before the original owner of the account does and post about the event, propagating the information. For the owner of the hacked account, this time difference may be of great importance.

This thesis aims to create a program that is able to detect when an account is hacked, by investigating the reactions of other people about the observed account.