Popular social networking sites such as Facebook and Twitter have millions of third-party applications (e.g. games and single-sign-on) which are regularly used by hundreds of millions of users. Unfortunately, the abuse of third-party applications to launch reputation manipulation attacks (e.g. fake likes/followers), propagate spam/malware, and misuse user data (e.g. Cambridge Analytica Fiasco) is quite prevalent. My research focuses on systematically investigating such cases of third-party application abuse by performing longitudinal measurements and proposing data-drive countermeasures to mitigate this abuse.