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 ›› 2011, Vol. 37 ›› Issue (20): 154-156. doi: 10.3969/j.issn.1000-3428.2011.20.053

• 人工智能及识别技术 • 上一篇    下一篇

用于P2P流媒体识别的类增量学习算法

李 进,张 鑫,王 晖   

  1. (国防科学技术大学信息系统与管理学院,长沙 410073)
  • 收稿日期:2011-06-10 出版日期:2011-10-20 发布日期:2011-10-20
  • 作者简介:李 进(1982-),男,硕士研究生,主研方向:多媒体网络,P2P流量检测;张 鑫,讲师、博士;王 晖,教授、博士生导师
  • 基金资助:
    国家“863”计划基金资助项目“非结构化P2P视频组播流实时识别与过滤技术”(2008AA01Z407)

Class Incremental Learning Algorithm for P2P Streaming Media Identification

LI Jin, ZHANG Xin, WANG Hui   

  1. (College of Information System & Management, National University of Defense Technology, Changsha 410073, China)
  • Received:2011-06-10 Online:2011-10-20 Published:2011-10-20

摘要: 针对P2P流媒体流量识别中的类增量学习问题,提出一种基于“一对一”支持向量机多分类器的类增量学习算法CIOOL。充分利用原有多分类器知识,在不打破原有分类器体系的前提下加入新增类样本知识,以构造出新的多分类器。实验结果表明,CIOOL算法能在保证识别精度的同时减少训练时间和内存消耗,是一种解决P2P流媒体流量识别中类增量问题的有效方法。

关键词: P2P流媒体识别, 类增量学习, 一对一, 支持向量机, CIOOL算法

Abstract: This paper studies class incremental learning of P2P streaming traffic identification by using one-against-one Support Vector Machine(SVM) multi-classification. A new SVM class incremental learning algorithm——Class Incremental One-against-One Learning(CIOOL) is presented. CIOOL can adequately use former knowledge to construct a new multi-classifier without training over again. Experimental results indicate that CIOOL can decrease the time of training and memory consuming, and it is an effective algorithm to solve the problem of class incremental learning in P2P streaming traffic identification.

Key words: P2P streaming media identification, class incremental learning, one-against-one, Support Vector Machine(SVM), CIOOL algorithm

中图分类号: