计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (2): 151-155.

• 数据库、数据挖掘、机器学习 • 上一篇    下一篇

人工蜂群算法优化支持向量机的分类研究

李璟民,郭  敏   

  1. 陕西师范大学 计算机科学学院,西安 710062
  • 出版日期:2015-01-15 发布日期:2015-01-12

Study on classification of artificial bee colony algorithm to optimization of support vector machine

LI Jingmin, GUO Min   

  1. School of Computer Science, Shaanxi Normal University, Xi’an 710062, China
  • Online:2015-01-15 Published:2015-01-12

摘要: 为了提高支持向量机分类准确率,采用人工蜂群算法对支持向量机参数进行优化,并将该优化方法应用于小麦完好粒、霉变粒和发芽粒三类麦粒的识别。使用小波变换分解信号能量作为特征向量,以分类错误率的倒数作为适应度函数,利用人工蜂群算法对支持向量机的惩罚因子和核函数宽度参数进行优化,优化SVM方法对小麦完好粒、霉变粒和发芽粒的分类正确率达到86%以上。实验结果表明,该研究有较强的实用价值,为SVM性能优化提供了一种新的方法。

关键词: 人工蜂群算法, 支持向量机, 参数优化, 小麦碰撞声, 分类

Abstract: In order to improve the classification precision of Support Vector Machines(SVM), a parameter optimization method based on artificial bee colony algorithm is proposed to solve this problem. Then the proposed method is applied to the recognition of undamaged wheat kernel, moldy damaged wheat kernel and sprout-damaged wheat kernel. The wavelet transform is applied to wheat impact acoustic signals and the normalized energy values of every frequency band are extracted to compose feature vectors. And the inverse of classification error rate is used as fitness value, and the artificial bee colony algorithm is used to optimize the penalty factor and kernel parameter of SVM. Then the optimized SVM is used to classify the undamaged kernel, moldy damaged kernel and sprout-damaged kernel, and the recognition accuracy rate is above 86%. The experimental results show that the research has a more universal value in application and provide a novel method for the optimization of SVM performance as well.

Key words: artificial bee colony algorithm, Support Vector Machine(SVM), parameter optimization, wheat impact acoustic signals, classification