一种基于自适应遗传策略的特征选择算法  

A feature selection method based on adaptive genetic strategy

作　　者：张云鹏[1] 闫一功[1] 

机构地区：[1]长春工业大学基础科学学院,吉林长春130012

出　　处：《长春工业大学学报》2010年第2期126-131,共6页Journal of Changchun University of Technology

基　　金：吉林省教育厅科研计划基金资助项目(吉教科合字[2004]第17号)

摘　　要：以欧氏相似测度散布矩阵为特征评价准则,以特征向量为基因构造二进制染色体码串,研究了一种基于自适应交叉/变异算子模型的遗传搜索算法。测试结果表明,对于解域空间具有非连续、多峰、含噪声特征的一类特征选择问题,该算法具有很强的鲁棒性,且能够以满意的概率收敛于全局满意解。With euclidean similarity scatter matrix as evaluation criteria, the binary chromosome code string is constructed with feature vectors as gene. We study a genetic scouting algorithm based on the adaptive crossover/mutation operator model, and the results shows the algorithm has a strong robustness and is able to converge to the global optimal solution with a satisfactory probability to the feature selection problem when the solution domain space is a non-continuous, multi-peak and with noises

关 键 词：特征提取 遗传算法 自适应 汽车发动机 故障诊断 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]