Improving the Result of Personalized Questionnaire Towards Solving Cold User Problem

Abstract

Collaborative filtering techniques is among the popular approaches used in addressing product recommender systems, which uses ratings and predictions to make new suggestions. However the major weakness of collaborative filtering approaches is cold user problem. Literature investigation has shown that cold user problem could be effectively addressed using active learning technique of administering personalized questionnaire. Unfortunately, the result of personalized questionnaire technique could contain some user preference uncertainties where the product database is too large (as in Amazon.com). This research work tends to address the weakness of personalized questionnaire technique by applying the active learning technique of uncertainty reduction over the result obtained from administering personalized questionnaire. This strategy has the tendency of resolving user preference uncertainties that could be associated with the result of personalized questionnaire. This research work is in progress. Preliminary result is encouraging.