Bhattacharyya distance feature selection
暂无分享,去创建一个
A recursive algorithm named Bhattacharyya distance feature selection for selecting a real-optimum feature under normal multidistribution is presented. The key of this method is to change the problem of minimizing the criterion of sum of the upper bound of error probability of every two class pair in subspace to a problem of solving nonlinear matrix equation in multiclass problem under orthonormal coordinate system. The recursive algorithm is considered as finding the optimal solution of transformation matrix from the nonlinear matrix equation. The theoretical analysis and experimental results show that under normal multidistribution the performance of proposed algorithm is superior to the performance of any previous one .
[1] Thomas Marill,et al. On the effectiveness of receptors in recognition systems , 1963, IEEE Trans. Inf. Theory.
[2] John W. Sammon,et al. An Optimal Set of Discriminant Vectors , 1975, IEEE Transactions on Computers.
[3] M. Thomason. Interactive Pattern Recognition , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.
[4] Shingo Tomita,et al. An optimal orthonormal system for discriminant analysis , 1985, Pattern Recognit..