TY  - CONF
ID  - www2009122
UR  - http://www2009.eprints.org/122/
A1  - Zhang, Congle
A1  - Xue, Gui-Rong
A1  - Yu, Yong
A1  - Zha, Hongyuan
Y1  - 2009/04//
N2  - Traditional Naive Bayes Classi?er performs miserably on web-scale taxonomies. In this paper, we investigate the reasons behind such bad performance. We discover that the low performance are not completely caused by the intrinsic limitations of Naive Bayes, but mainly comes from two largely ignored problems: contradiction pair problem and discriminative evidence cancelation problem. We propose modi?cations that can alleviate the two problems while preserving the advantages of Naive Bayes. The experimental results show our modi?ed Naive Bayes can signi?cantly improve the performance on real web-scale taxonomies. 
TI  - Web-Scale Classification with Naive Bayes
SP  - 1083
M2  - Madrid, Spain
AV  - public
EP  - 1083
T2  - 18th International World Wide Web Conference
ER  -