Abstract

Stereotypes and clustering are some techniques for creating user models from user behavior. Yet, they possess important risks as users actions could be misinterpreted or users could be associated with undesirable profiles. It could be worst if users' actions, beliefs, and comments are long term stored such as in Organizational Memory Systems (OMS) where users' contributions are available to the whole organization. We propose a privacy model based on four privacy roles that allow users to control the disclosure of their personal data and, when recovered, blurs such data as time passes. © Springer-Verlag Berlin Heidelberg 2005.