基于改进的和声搜索算法的特征基因选择方法

摘要

DNA microarray data often contain tens of thousands of genes,where there are a lot of irrelevant and redundant genes. These genes may seriously affect the accuracy and efficiency of classification. In order to solve this problem. This paper proposes a feature gene selection method based on improved harmony search algorithm. Firstly , microarray genes are ranked using ReliefF algorithm and preselected gene subset is obtained according to ranked⁃top genes,then the improved harmony search algorithm is used to select feature genes from above gene subset. Finally we implement simulation experiments on three public microarray data sets. The results show that the proposed algorithm can achieve very high accuracy in less feature genes,and is a effective feature gene selection algorithm.%DNA微阵列数据通常含有成千上万个基因，其中含有大量与分类无关的基因和冗余基因，这些基因的存在会严重影响分类精度和效率。针对这一问题，提出一种基于改进的和声搜索算法的特征基因选择方法，首先采用ReliefF算法对微阵列基因数据集排序，取排序靠前的N个基因构成初选基因子集，然后再利用改进的和声搜索算法选择特征基因。通过在3个公共微阵列数据集上的仿真实验，结果表明，该算法能够在更少的特征基因情况下达到很高的精度，是一种有效的特征基因选择算法。