计算机应用 ›› 2012, Vol. 32 ›› Issue (06): 1522-1525.DOI: 10.3724/SP.J.1087.2012.01522

• 网络与通信 • 上一篇    下一篇

基于语义描述与优化的网络性能数据聚类方法

姜大庆1,2,周勇2,夏士雄2   

  1. 1. 南通农业职业技术学院 信息工程系,江苏 南通 226007
    2. 中国矿业大学 计算机科学与技术学院,江苏 徐州 221008
  • 收稿日期:2011-11-25 修回日期:2012-02-02 发布日期:2012-06-04 出版日期:2012-06-01
  • 通讯作者: 姜大庆
  • 作者简介:姜大庆(1969-),男,江苏南通人,副教授,硕士,CCF会员,主要研究方向:数据挖掘、Web应用;〓周勇(1974-),男,江苏徐州人,副教授,博士,主要研究方向:数据挖掘、智能信息处理;〓夏士雄(1961-),男,黑龙江鹤岗人,教授,博士生导师,主要研究方向:模式识别、人工智能。
  • 基金资助:
    国家自然科学基金资助项目;江苏省教育厅“青蓝工程”基金资助;南通市科技产业化计划项目

Network performance data clustering method based on semantic description and optimization

JIANG Da-qing1,2,ZHOU Yong2,XIA Shi-xiong2   

  1. 1. Department of Information Engineering, Nantong Agricultural College, Nantong Jiangsu 226007, China
    2. School of Computer Science and Technology, China University of Mining and Technology, Xuzhou Jiangsu 221008,China
  • Received:2011-11-25 Revised:2012-02-02 Online:2012-06-04 Published:2012-06-01
  • Contact: JIANG Da-qing

摘要: 为了从多源复杂的网络性能数据中挖掘有用模式以提高网络服务质量,研究了基于本体的网络性能监测数据聚类分析方法。阐述了网络性能监测数据的语义描述方法,提出基于语义和属性数据相融合的网络性能数据相似性度量模型,并给出基于改进k-means的NJW谱聚类算法。通过在UCI数据集和校园网性能监测数据集上的实验表明, 本文所提方法较相关比对方法具有更高的聚类准确性和区分度。

关键词: 本体, 语义描述, 语义相似度, 谱聚类, 网络性能监测

Abstract: In order to improving the network quality of service by mining useful model from multi-source and complicated network performance data, a clustering analysis algorithm for network performance monitoring data based on ontology. The semantic description method of network performance monitoring data is described, then a similarity measurement model of network performance data based on semantic description and property data is proposed, and an NJW spectral clustering algorithm based on improved k-means algorithm is given. The experiments based on the UCI data sets and the performance monitoring data on a campus network shows that the proposed algorithm has a higher clustering accuracy and differentiation than the comparative algorithms.

Key words: ontology, semantic description, semantic similarity, spectral clustering, network performance monitoring