Authorship Attribution in Health Forums

The emergence of social media (networks, blogs, web forums) has given people numerous opportunities to share their personal stories, including details of their health. Although users mostly post under assumed nicknames, state-of-the-art text analysis techniques can combine texts from different media and use that linkage to identify private details of an individual‟s health. In this study we aim to empirically examine the accuracy of identifying authors of on-line posts on a medical forum. 1 Our results show a high accuracy of the authorship attribution, especially when text is represented by the orthographical features.