Adapt the mRMR Criterion for Unsupervised Feature Selection
Feature selection is an important task in data analysis. mRMR is an equivalent form of the maximal statistical dependency criterion based on mutual information for first-order incremental supervised feature selection. This paper presents a novel feature selection criterion which can be considered as the unsupervised version of mRMR. The concepts of relevance and redundancy are both concerned in the feature selection criterion. The effectiveness of the new unsupervised feature selection criterion is confirmed by the theoretical proof. Experimental validation is also conducted on several popular data sets, and the results show that the new criterion can select features highly correlated with the latent class variable.
Feature selection Unsupervised feature selection Mutual information
Junling Xu
School of Computer Science and Engineering,Southeast University, Nanjing 210096, China
国际会议
6th International Conference on Advanced Data Mining and Applications(第六届先进数据挖掘及应用国际会议 ADMA 2010)
重庆
英文
111-121
2010-11-19(万方平台首次上网日期,不代表论文的发表时间)