An method of self-learning in adaptive text information filtering
Hui Ying Ning, Tan Ya-zhou, LV Zhi-long, Yue Wu, Cui Li-gang, Wang Chun-hua · 2010
With the growing popularity of Internet and development, the amount of information on the Internet is beyond people's imagination. Information filtering is searching interesting information for user and shielding other useless information in dynamic information flow based on user's need of information. In this background, information filtration comes into being and becomes a very important branch in information processing field. Because text information is one of the most important forms of Internet information, this article focuses on text filtering processing. This paper mainly searches how to build user template more precisely, the template learning algorithm in filtering processing and adaptive threshold in adaptive information filtering processing based on vector space model. Adaptive learning is fulfilled through genetic optimization of feedback information. Updating user templates by data gained by adaptive learning helps adaptive filtering. According to the experimental result, this method has shielded the information sparsely of the pseudo-relevance feedback and the misleading of the feature ambiguity effectively to improve the filtering quality of the adaptive information filtering system.