Mining users' two-dimension interests from cache

Abstract

Popular WWW pages are stored in places which are close to users by WWW cache technology to speed up the fetching of these objects. Information in the WWW cache shows users' recent interest. Users' interest can be widely used in customizing WWW pages, filtering information, pre-fetching information and so on. The key to using information in the WWW cache effectively is to build an adaptive user interest model and construct an adaptive algorithm for interest mining. Interest can be specialized by the tuple (term, weight) in the simple interest model. Association relations are not mined in this model, so the interest cannot be associated in expressing users' interest. Basing on analyzing the WWW cache model, we propose a two-dimension interest model. It is proved that mined users' interest can show users' current interest states. Inferential relations are well considered in the two-dimension interest model. This model is not a simple extension of the simple interest model, but the full update and the related algorithm. Interrelated methods for storing the two-dimension interest model, computing the model effectively and updating the model in real time are considered.