先按照相关性系数标准对样本所包含的全体基因进行筛选,降低冗余的同时有利于缩小优化算法的搜索范围;然后在筛选结果上采用蚁群优化策略进行分类相关基因子集的选择,并利用样本聚类效果作为优化目标函数,在保证分类准确度的同时大大降低了基因选择方法的计算复杂度。实验结果表明,提出的基因选择方法能在相当短的时间内选出与分类相关的基因子集。 Firstly, the whole gene contained in the sample is screened according to the correlation coefficient standard to reduce the redundancy and reduce the searching scope of the optimization algorithm. Then, the ant colony optimization strategy is used to select the subset of related genes Using the sample clustering effect as the optimization objective function, the computational complexity of the gene selection method is greatly reduced while ensuring the classification accuracy. The experimental results show that the proposed gene selection method can select gene subset related to classification in a relatively short period of time.