Feature selection based on quadratic mutual information criterion

Quadratic mutual information was used as the criterion for feature variable selection in pattern recognition problems,and the geometric meaning of QMI criterion in the reproducing kernel Hilbert space was analyzed. Based on the criterion,a new feature selection algorithm,PW-QMI,was proposed,which used Parzen window for probability density estimation and backward elimination strategy for searching in feature variable space.In situations of a large size example set,another algorithm,GMM-QMI,was proposed which used the Gaussian mixture model for density estimation to reduce computation complexity.The comparative experiments on the correlation-criterion-based algorithm and SVM-RFE algorithm show the stable performance of the proposed algorithms for feature selection.