OCFS: Optimal Orthogonal Centroid Feature Selection for Text Categorization

Jun Yan, Ning Liu, Benyu Zhang, Shuicheng Yan, Zheng Chen, Qiansheng Cheng, Weiguo Fan, Wei-Ying Ma. Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval. Experiments based on 20 Newsgroups (20NG), Reuters Corpus Volume 1 (RCV1) and ODP data show that OCFS is a consistently better feature selection method than Information Gain (IG) and c 2-test (CHI).