Title :
News Contents Recommendation Model Based on Feedback of Web Usage
Author :
Ping Ni ; Liao, Jianxin ; Zhu, Xiaomin ; Ren, Keyan
Author_Institution :
State Key Lab. of Networking & Switching Technol., Beijing Univ. of Posts & Telecommun., Beijing, China
fDate :
March 31 2009-April 2 2009
Abstract :
In this paper, reclassification for the current classification through K-means would be implemented based on the feedback of Web usage mining in order to improve the accuracy of news recommendation and convergence of classification. It could extract most relative keywords and eliminate the disturbance of multi-vocal word in one category based on feedback of Web usage. The reclassification of news contents would be implemented based on K-means algorithm and Web usage mining result. We call this method as ReK-means. By simulation comparing, accuracy of reclassification were obvious to be improved compared with related words classification algorithm.
Keywords :
Web sites; classification; data mining; document handling; information retrieval; K-means classification; ReK-means method; Web document categorization; Web site usage mining feedback; keyword extraction; multivocal word; news content reclassification; news content recommendation model; related words classification algorithm; Automation; Clustering algorithms; Computer science; Data mining; Electronics industry; Information technology; Laboratories; State feedback; Telecommunication switching; Web server; machine learning; news recommendation; web mining;
Conference_Titel :
Computer Science and Information Engineering, 2009 WRI World Congress on
Conference_Location :
Los Angeles, CA
Print_ISBN :
978-0-7695-3507-4
DOI :
10.1109/CSIE.2009.104
