作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2009, Vol. 35 ›› Issue (17): 184-186. doi: 10.3969/j.issn.1000-3428.2009.17.063

• 人工智能及识别技术 • 上一篇    下一篇

改进的基于K均值聚类的SVDD学习算法

花小朋,李先锋,皋 军,田 明   

  1. (盐城工学院信息工程学院,盐城 224001)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-09-05 发布日期:2009-09-05

Updated Learning Algorithm of Support Vector Data Description Based on K-Means Clustering

HUA Xiao-peng, LI Xian-feng, GAO Jun, TIAN Ming   

  1. (School of Information Technology, Yancheng Institute of Technology, Yancheng 224001)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-09-05 Published:2009-09-05

摘要: 针对基于K均值聚类的支持向量数据描述(SVDD)学习算法(KMSVDD)识别精度低于传统SVDD学习算法的问题,提出一种改进算法。将各聚类簇中支持向量合并学习生成中间模型,从支持向量以外的非支持向量数据中找出违背中间模型KKT条件的学习数据,并将这些数据与聚类簇中支持向量合并学习继而得到最终学习模型。实验结果证明,该改进算法的计算开销与KMSVDD相近,但识别精度却高于KMSVDD,与传统SVDD相近。

关键词: 支持向量数据描述, K均值, KKT条件

Abstract: Aiming at the flaw that the recognition precision of Support Vector Data Description based on K-Means(KMSVDD) clustering is lower than traditional Support Vector Data Description(SVDD), an improvement algorithm is proposed. This algorithm learns support vectors of every cluster and produces middle model, then finds out the data against middle model’s Karush-Kuhn-Tucker(KKT) condition from non-support vectors and obtains the final studying model by leaning them with all support vectors. Experimental result proves that this improvement algorithm has similar computing expenditure with KMSVDD and its recognizing accuracy is higher than KMSVDD and similar to traditional SVDD.

Key words: Support Vector Data Description(SVDD), K-Means, Karush-Kuhn-Tucker(KKT) condition

中图分类号: