计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (19): 31-34.

• 研究、探讨 • 上一篇    下一篇

基于动态数据流挖掘的案例推理及其应用

戴奇波1,2,倪志伟1,2,王 超1,2,姜 苗1,2   

  1. 1.合肥工业大学 管理学院,合肥 230009
    2.过程优化与智能决策教育部重点实验室,合肥 230009
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-07-01 发布日期:2011-07-01

Case based reasoning and application based on dynamic data stream mining

DAI Qibo1,2,NI Zhiwei1,2,WANG Chao1,2,JIANG Miao1,2   

  1. 1.School of Management,Hefei University of Technology,Hefei 230009,China
    2.Key Laboratory of Process Optimization and Intelligent Decision-making,Ministry of Education,Heifei 230009,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-07-01 Published:2011-07-01

摘要: 知识的获取、知识库的更新是案例推理技术的应用瓶颈,而许多案例推理系统中的知识库都是静态不变的,满足不了实际问题变化的需要。首先阐述了相关概念,接着提出了一种基于动态数据流挖掘的案例推理模型,其中动态数据流挖掘算法采用改进的数据流聚类算法。通过此模型使用基于动态数据流挖掘的案例推理技术,对数据进行实时挖掘,产生连续、动态的临时案例库,实现知识库的实时更新,从而满足实际问题变化的需要。最后通过该模型在实际中的应用说明其有效性。

关键词: 数据流, 案例推理, 聚类

Abstract: The application of case-based reasoning is restricted by the knowledge acquisition and the knowledge base updating.Many knowledge bases in the case-based reasoning system are static and unchangeable,and can not satisfy the change of practical problems.This paper describes the relevant concepts and presents a model of CBR based on dynamic data stream mining,and gives an improved clustering algorithm of data stream.Through this model the system can mine real-time datum,produce continuous,dynamic temporary cases,update the knowledge base in real time and meet the needs of the practical problems.Finally,the application of the model in practice verifies its efficiency.

Key words: data stream, case based reasoning, clustering