一种基于PSO同步进行特征选择及参数优化的核K近邻分类算法

Kernel KNN Algorithm based on Simultaneous Feature Selection and Parameter Optimization by PSO

特征选择及分类器参数优化是提高分类器性能的两个重要方面,传统上这两个问题是分开解决的.近年来,随着进化优化计算技术在模式识别领域的广泛应用,编码上的灵活性使得特征选择及参数的同步优化成为一种可能和趋势.为了解决此问题,本文研究采用二进制PSO算法进行特征选择及核K近邻分类器参数的同步优化.实验表明,该方法可有效地找出合适的特征子集及核函数参数,并取得较好的分类效果.

Feature selection and classifier parameter optimization are two important aspects for improving classifier performance and are solved separately traditionally. Recently, with the wide applications of evolutionary computation in pattern recognition area, simultaneous feature selection and parameter optimization became possible and tendency. To solve the problem, we proposed a simultaneous feature selection and parameter optimization algorithm based on binary PSO algorithm. The experiments show that the algorithm can efficiently find the suitable feature subsets and parameters, which result in good classification performance.