J4 ›› 2012, Vol. 50 ›› Issue (06): 1199-1203.

Previous Articles     Next Articles

An Improved Web Structure Similarity Based on MatchingAlgorithm of Tree Paths

LIAO Haowei, YANG Yan, JIA Zhen, YIN Hongfeng   

  1. School of Information Science and Technology, Southwest Jiaotong University, Chengdu 610031, China
  • Received:2012-05-21 Online:2012-11-26 Published:2012-11-26
  • Contact: YANG Yan E-mail:yyang@swjtu.edu.cn

Abstract:

An improved algorithm of Web structure similarity based on tree path matching was proposed, which defines the sequence similarity and position similarity of the tree path, finds out all the Web tree paths, and calculates the structural similarity by best tree path matching between two Web pages. Experiments show that the proposed algorithm to calculate the Web structure similarity is more realistic and effective than the original algorithm.

Key words: Web structure similarity, sequence similarity, position similarity

CLC Number: 

  • TP391