LEARNING AND REVISING USER PROFILES - THE IDENTIFICATION OF INTERESTING WEB SITES

We discuss algorithms for learning and revising user profiles that can determine which World Wide Web sites on a given topic would be intere sting to a user. We describe the use of a naive Bayesian classifier fo r this task, and demonstrate that it can incrementally learn profiles from user feedback on the interestingness of Web sites. Furthermore, t he Bayesian classifier may easily be extended to revise user provided profiles. In an experimental evaluation we compare the Bayesian classi fier to computationally more intensive alternatives, and show that it performs at least as well as these approaches throughout a range of di fferent domains. In addition, we empirically analyze the effects of pr oviding the classifier with background knowledge in form of user defin ed profiles and examine the use of lexical knowledge for feature selec tion. We find that both approaches can substantially increase the pred iction accuracy.