“You Are What You Write”--Understanding User Online Behavior through Text Mining

Recently, thanks to Web 2.0 technologies that make it easy for users to contribute content to online communities, there is an explosion of user participation online. As more users contribute content, the digital trace they leave offers huge amount of data for studying their behavior. For example, as a relative new social network site, Facebook allows users to set up their own page, establish relationships, communicate and share content and applications with other users online. Even many non-social network web sites have added community features such as commenting, user management, and deep profiling, private messaging, and friends.Such explosive growth offers new opportunities for studying online behavior. Three types of data can be analyzed to offer a complete view of users, namely, text data, user actions, and social network data. In other words, we can now observe what people say, what they do, as well as who they make friends with. Together they allow researchers to have a more complete understanding of user behavior online.