J4 ›› 2012, Vol. 39 ›› Issue (3): 126-130.doi: 10.3969/j.issn.1001-2400.2012.03.020

• 研究论文 • 上一篇    下一篇

潜在语义分析的供求信息自动匹配算法

冯月进;张凤斌   

  1. (哈尔滨理工大学 计算机科学与技术学院,黑龙江 哈尔滨  150080)
  • 收稿日期:2011-01-12 出版日期:2012-06-20 发布日期:2012-07-03
  • 通讯作者: 冯月进
  • 作者简介:冯月进(1970-),男,哈尔滨理工大学博士研究生,E-mail: yjfeng@hotmail.com.
  • 基金资助:

    国家自然科学基金资助项目(60671049)

Automatic matching algorithm for the latent semantic analysis based supply and demand information

FENG Yuejin;ZHANG Fengbin   

  1. (Computer Sci. and Tech. Inst., Harbin Univ. of Sci. and Tech., Harbin  150080, China)
  • Received:2011-01-12 Online:2012-06-20 Published:2012-07-03
  • Contact: FENG Yuejin

摘要:

将潜在语义分析应用于电子商务系统的供求信息匹配中,解决了传统模型中同义和多义现象对匹配精度有很大负面影响的问题;同时通过引入信息熵,改进了潜在语义分析的权重计算,提出了基于潜在语义分析的、结合了规则提取和相关反馈的供求信息自动匹配算法,并给出了配套的供求信息规则库的设计方法.实验结果显示,该算法具有很好的匹配精度,性能明显优于基于空间向量模型的供求信息匹配方法.

关键词: 潜在语义分析, 信息熵, 语义, 供求信息匹配, 向量空间模型

Abstract:

In traditional Supply and Demand Information Matching models, a word is regarded as an independent unit. However, there are many synonyms and polysemy in the natural language and their existence has deteriorated the precision. In order to solve this problem, Latent Semantic Analysis is applied to it. Moreover, an algorithm based on Entropy is proposed to improve the weighting of Latent Semantic Analysis. A Supply and Demand Information Automatic Matching algorithm based on Latent Semantic Analysis, Rule Extraction and Relevance Feedback is realized. And a Supply and Demand Information Base is designed to support it. Experimental results show that the precision of this algorithm is much better than that of the method based on the Vector Space Model.

Key words: latent semantic analysis, entropy, semantic, supply and demand information matching, vector space model