SUMMARY OF LASSO AND RELATIVE METHODS

摘要：

Feature Selection is one of the focuses in pattern recognition field. To select the most obvious features, there are some feature selection methods such as LASSO, Bridge Regression and so on. But all of them are limited in select feature. In this paper, a summary is listed. And also the advantages and limitations of every method are listed. By the end, an example of LASSO using in identification of Traditional Chinese Medicine is introduced to show how to use these methods to select the feature.

关键词： Feature selection Variable selection Pattern recognition LASSO Ridge regression

作者: Xia Jianan Sun Dongyi Xiao Fan

作者单位: Department of Information and Computational Science, Beijing Jiaotong University, Beijing, China

会议类型: 国际会议

会议名称: 13th International Conference on Enterprise Information System(第13届企业信息系统国际会议 ICEIS 2011)

会议地点: 北京

会议语种:英文

页码: 2598-2601

在线出版日期: 2011-06-08（万方平台首次上网日期，不代表论文的发表时间）