A new semantic similarity measuring method based on web search engines

  • Authors:
  • Gang Lu;Peng Huang;Lijun He;Changyong Cu;Xiaobo Li

  • Affiliations:
  • School of Information, Zhejiang University of Finance & Economics, Hangzhou, China;Zhejiang University, Hangzhou, China;School of Information, Zhejiang University of Finance & Economics, Hangzhou, China;School of Mechanical Engineering, Hangzhou Dianzi University, Hangzhou, China;School of Information, Zhejiang Education Institute, Hangzhou, China

  • Venue:
  • WSEAS Transactions on Computers
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Word semantic similarity measurement is a basic research area in the fields of natural language processing, intelligent retrieval, document clustering, document classification, automatic question answering, word sense disambiguation, machine translation, etc.. To address the issues existing in current approaches to word semantic similarity measurement, such as the low term coverage and difficult update, a novel word semantic similarity measurement method based on web search engines is proposed, which exploits the information, including page count and snippets, in retrieved results to do calculation. The proposed method can resolve the issues mentioned above due to the huge volumes of information in the Web. The experimental results demonstrate the effectiveness of the proposed methods.