基于免疫克隆选择算法的特征选择

Feature Selection Based on Immune Clonal Selection Algorithm

提出了一种基于免疫克隆选择算法的特征选择方法.特征选择可以被看成是一个组合优化问题,利用免疫克隆选择算法快速收敛于全局最优的特性,加快搜索到最优特征子集的速度,为后续模式分类提供良好的判别依据.实验结果表明算法在保持甚至提高分类精度的同时,有效地降低了特征维数.与基于遗传算法特征选择的结果相比较,在有限代数内,该算法能收敛到更优的特征子集,从而验证了算法的有效性及其应用潜力.

A new feature selection algorithm based on Immune Clonal Selection Algorithm(ICSA)is proposed. Feature selection can be considered as an optimization problem. The property of rapid convergence to global optimum of ICSA is made use of to speed up the searching of the most suitable feature subset among a huge number of possible feature combinations. The experimental results show that this approach can efficiently reduce the number of features while maintaining or even improving the accuracy. In addition, compared with Genetic Algorithm (GA) based feature selection, the proposed method can find better feature subset for classification in the limited number of evolutionary generations. Accordingly, the high effectiveness and great potential of the new method are demonstrated.