Title :
TC-PageRank Algorithm Based on Topic Correlation
Author :
Huang, Decai ; Qi, Huachun ; Yuan, Yuan ; Zheng, Yue-feng
Author_Institution :
Coll. of Inf. Eng., Zhejiang Univ. of Technol., Hangzhou
Abstract :
PageRank algorithm is a famous algorithm to mine the Web structure, but it has a drawback of topic-drift. To eliminate the topic-drift of the PageRank algorithm, and after the analysis of existing algorithms, a new algorithm called TC-PageRank algorithm is put forward. The TC-PageRank algorithm is based on fictitious file vector and correlation measure of cosine. Experimental results illustrate that TC-PageRank algorithm eliminates the topic-drift phenomenon effectively, and thus improves the quality of retrieving
Keywords :
Internet; correlation methods; data mining; information retrieval; TC-PageRank; Web structure mining; cosine correlation measure; fictitious file vector; topic correlation; topic-drift drawback; Algorithm design and analysis; Automobiles; Classification algorithms; Educational institutions; Internet; Prototypes; Search engines; Turning; Web sites; Hyperlink Analysis; PageRank Algorithm; Topic Correlation; Web Structure Mining;
Conference_Titel :
Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on
Conference_Location :
Dalian
Print_ISBN :
1-4244-0332-4
DOI :
10.1109/WCICA.2006.1714219
