基于混沌机制的人工蜂群算法优化的支持向量机分类器

支持向量机的分类性能在很大程度上取决于其相关参数的选择，为了改善支持向量机的分类准确率，本文采用基于混沌机制的人工蜂群算法对其参数进行优化。在传统人工蜂群算法的基础上，采用Logistic混沌映射初始化种群和锦标赛选择策略，进一步提高人工蜂群算法的收敛速度和寻优精度。该方法采用分类准确率作为适应度函数，利用人工蜂群算法对支持向量机的惩罚因子和核函数参数进行优化。通过对多个标准数据集的分类测试，证明基于混沌机制的人工蜂群算法优化的支持向量机分类器能够获得更高的分类准确率。

Artificial Colony Algorithm Based on Chaotic Mechanism Optimization of Support Vector Machine Classifier

The classification performance of support vector machine (SVM) to a large extent depends on the selection of its parameters, so this paper used artificial bee colony algorithm based on chaotic mechanism to optimize the parameters in order to improve the classification accuracy of support vector machine (SVM). On the basis of the traditional artificial colony algorithm, by using the Logistic chaotic mapping initialization population and tournament selection strategy, the artificial colony algorithm convergence speed and optimization precision can be further improved. The method adopts the classification accuracy as fitness function, and uses artificial colony algorithm of support vector machine (SVM) penalty factor and the kernel function parameter optimization. By standard data sets with the classification of the test, it proves that artificial colony algorithm based on chaotic mechanism optimization of support vector machine classifier can achieve higher classification accuracy.