Evolving Information Filtering for Personalized Information Service

Information filtering (IF) systems areimportant for personalized information service. However, most current IFsystems suffer from low quality and long training time. In this paper,a refined evolving information filtering method is presented. Thismethod describes user's information need from multi-aspects and improvesfiltering quality through a process like natural selection. Experimentalresult shows this method can shorten training time, improve filteringquality, and reduce the relevance between filtering results and trainingsequence.