Incremental Feature Selection for Online Classification

摘要：

The contribution of this paper is in two-fold. First, Incremental feature selection based on correlation ranking (CR) is proposed for classification problem. Second, the performance of combining Random forest (RF) learning algorithm with the proposed feature selection strategies is investigated. We show that such form of feature selection will be useful towards on-line classifier. Evaluation based on the dataset of NIPS 2003 feature selection challenge demonstrates that our method can achieve comparable performance.

关键词： Index Terms—Feature selection on-line learning random

作者: HASSAB ELGAWI Osman

作者单位: Computational Intelligence and Systems Science,Tokyo Institute of Technology-Japan

会议类型: 国际会议

会议名称: 2008高等智能国际会议(2008 International Conference on Advanced Intelligence)

会议地点: 北京

会议语种:英文

在线出版日期: 2008-10-18（万方平台首次上网日期，不代表论文的发表时间）