Metric in Feature Space
In this paper our purpose is to define a metric in feature space. The metric finds the correlation distance between two features. It may be used in many applications. An application showing the utility of the proposed metric for feature selection is described here. The performance of the feature selection method is found to be comparable to other feature selection methods.
Keywordsmetric correlation PCA partition entropy regression
- 1.Liu, H., Motoda, H.: Computational Methods of Feature Selection. Chapman and Hall/Crc Data Mining and Knowledge Discovery Series (2007)Google Scholar
- 7.Blake, C.L., Merz, C.J.: UCI Repository of Machine Learning Database, Univ. of California, Irvine, Dept. of Information and Computer Science (1998)Google Scholar