Parameters optimization of classifier and feature selection based on improved artificial bee colony algorithm
Haiquan Wang, Hongnian Yu, Qian Zhang, Shuang Cang, Wudai Liao, Fanbing Zhu · 2016
The feature subset selection, along with the parameters of classifier significantly influences the classification accuracy. In order to ensure the optimal classification performance, the artificial bee colony (ABC) algorithm is proposed to simultaneously optimize the feature subset and the parameters of support vector machines (SVM), meanwhile for improving the optimizing performance of ABC algorithm, the initialization and scout bee phase are improved. To evaluate the proposed approach, the simulation was executed based on datasets from the UCI database. The effectiveness of the proposed method is confirmed by simulation results.