正交基低冗余无监督特征选择法

借鉴基于正则回归的无监督并行正交基聚类特征选择法和最大互信息系数,提出正交基低冗余无监督特征选择法.该方法在正交基下选择具有判别能力的特征,可用最大互信息系数矩阵选择低冗余性的特征子集. 4个图像数据集上的实验结果表明:该方法选择的特征子集可以提高聚类准确率.

Orthogonal basis minimization redundancy unsupervised feature selection method

Feature selection method plays an important role in improving the readability and reducing the complexity of data. Based on unsupervised simultaneous orthogonal basis clustering feature selection (SOCFS) and maximum mutual information coefficient (MIC), an orthogonal basis minimization redundancy unsupervised feature selection method (OBMRFS) is proposed. The method selects features with discriminant ability under orthogonal basis and uses maximum mutual information coefficient matrix to select the feature subset with minimization redundancy. The experimental results on four standard datasets show that the feature subset selected by the proposed method can improve the clustering accuracy.