基于边界点的可分离性度量及特征选择

Separability Measure Based on Boundary Points and Feature Selection

提出了一种新的面向高维数据的特征选择方法,在特征子集搜索上采用遗传算法进行随机搜索,在特征子集评价上采用基于边界点的可分性度量作为评价指标及适应度。实验结果表明,该算法可有效地找出具有较好的可分离性的特征子集,从而实现降维并提高分类精度。

This paper proposes a new feature selection method for the high-dimensional data, which realizes the feature subset search by genetic algorithm, and the feature subset fitness is evaluated by the separability measure based on boundary points. The experiments show that the proposed algorithm can find out the feature subsets with good separability, which results in the low-dimensional data and the good classification accuracy.