Title :
Improvement of Feature Extraction in Web Page Classification
Author :
Jiao Lijuan ; Feng Liping
Author_Institution :
Dept. of Comput. Sci., Xinzhou Teachers Univ., Xinzhou, China
Abstract :
Mutual information formula is improved by using the hyperlink factor in this paper. Introduction of hyperlink elements of web pages can improve the classification accuracy in feature selection method based on mutual information and correlation by experiment, especially for those of strong. So the improvement is effective in web page classification.
Keywords :
Web sites; classification; feature extraction; Web page classification; feature extraction; feature selection; hyperlink factor; Computer science; Data mining; Electronic mail; Feature extraction; Frequency; Mutual information; Optimization methods; Relational databases; Text categorization; Web pages;
Conference_Titel :
e-Business and Information System Security (EBISS), 2010 2nd International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-5893-6
Electronic_ISBN :
978-1-4244-5895-0
DOI :
10.1109/EBISS.2010.5473682
