Learning about Users from Observation

Many approaches and systems for recommending information, goods, or other kinds of objects have been developed in recent years. In these systems, machine learning methods are often used that need training input to acquire a user interest profile. Such methods typically need positive and negative evidence of the user’s interests. To obtain both kinds of evidence, many systems make users rate relevant objects explicitly. Others merely observe the user’s behavior, which yields positive evidence only; in order to be able to apply the standard learning methods, these systems mostly use heuristics to also find negative evidence in observed behavior. In this paper, we present an approach for learning interest profiles from positive evidence only, as it is contained in observed user behavior. Thus, both the problem of interrupting the user for ratings and the problem of somewhat artificially determining negative evidence are avoided. A methodology for learning explicit user profiles and recommending interesting objects has been developed. It is used in the context of ELFI - a Web-based information system. The evaluation results are briefly described in this paper. Our current efforts revolve around further improvements of the methodology and its implementation for recommending interesting web pages to users of a web browser.