Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (8): 138-140.DOI: 10.3778/j.issn.1002-8331.2009.08.042

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Reconstruction of Web sessions using DFA

HUANG Jin-jing,ZHAO Lei,YANG Ji-wen   

  1. School of Computer Science and Technology,Soochow University,Suzhou,Jiangsu 215006,China
  • Received:2008-01-23 Revised:2008-03-31 Online:2009-03-11 Published:2009-03-11
  • Contact: HUANG Jin-jing

使用DFA的Web会话构造方法

黄金晶,赵 雷,杨季文   

  1. 苏州大学 计算机科学与技术学院,江苏 苏州 215006
  • 通讯作者: 黄金晶

Abstract: Session reconstruction is an important step of data preprocessing in Web usage mining.This paper applies DFA theory to sessions reconstruction,aiming at a section of users’ browsing log,to recognize sessions reconstruction by the states conversion of DFA.This method pays more attention to the continuity of pages and the true sequence browsed by users,which benefits for mining users’ browsing patterns.

Key words: Web usage mining, data processing, sessions reconstruction, Deterministic Finite Automata(DFA)

摘要: 会话识别是Web使用挖掘数据预处理中重要的一个环节。将确定的有限自动机(DFA)思想运用于会话构造,针对一段用户访问日志,通过DFA中各个状态间的转换,实现会话构造。该方法更多考虑页面之间的连续性,关注用户的实际访问序列,有利于后续的用户访问模式的挖掘。

关键词: Web使用挖掘, 数据预处理, 会话识别, 确定的有限自动机(DFA)