基于Fisher线性判别模型的文本特征选择算法_刘健.pdfVIP

　　 　　　　　　　 305 　　　　 　　JOURNAL OFNATIONAL UNIVERSITY OF DEFENSE TEC NOLOGY　　　　 Vol.30 No.5 2008 :1001-2486(2008)05-0135-04 Fisher 刘　健, 钱　猛, 张维明 ( , 　410073) 　:, , 。Fisher , , 。 , 、 , 。 :Fisher ;; :TP391　　:B A Fisher Linear Discriminant Model-Based Text Feature Selection Algorithm LIU Jian, QIAN Meng, Z ANG Wei-ming (College of Information System and Management, National Univ.of Defense Technology, Changsha 410073, China) Abstract:Dimension reducing is very important in VSM based text classification system.Feature selection is more suitable for text databecause of its efficiency.A new feature selection algorithm is proposed in this paper on the basis of Fisher linear discriminant model, which converts the solution process to feature optimization problem and avoids the complex matrix operations.The experiment shows that the new algorithm has good performance and is better than IG and C Imethod. Key words:fisher linear discriminant model;text classification;feature selection , 、 、。(Vector Space Model) [1] , , 。 [1-3] , , , 。, 。 Fisher , , , , [4] 。Fisher , , , , , , 。 1　Fisher Fisher , , , , , 。 c1, c2, …, cL , L 。Sa Sb :2008-01-12 : :(1975—), , , 。 　1　36　　　　　　　　　　　　　　　　　　 　　　　　　　　　　　　　　2008 5 L T S = E [(X -m )(X -m ) |c ] (1) a ∑ j j j j =1