作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2006, Vol. 32 ›› Issue (20): 10-12. doi: 10.3969/j.issn.1000-3428.2006.20.004

• 博士论文 • 上一篇    下一篇

用户兴趣优化过滤方法研究

李 钝1,2,曹元大1,张龙飞1   

  1. (1. 北京理工大学计算机学院,北京 100081;2. 山西大学现代教育技术中心,太原 030006)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2006-10-20 发布日期:2006-10-20

New Optimized and Filtering Method of User’s Interest

LI Dun1,2, CAO Yuanda1, ZHANG Longfei1   

  1. (1. School of Computer, Beijing Institute of Technology, Beijing 100081; 2. Modern Education Technology Center, Shanxi University, Taiyuan 030006)

  • Received:1900-01-01 Revised:1900-01-01 Online:2006-10-20 Published:2006-10-20

摘要: 为了更准确地描述用户兴趣以及帮助他们得到真正感兴趣的信息,该文提出了一种优化加权用户兴趣进行过滤的方法。基于用户提出的过滤关键词的顺序赋予各关键词不同的权重,利用粗糙集理论对该用户兴趣进行同义优化回归过滤,得到用户所需要的信息。实验表明该方法可提高信息过滤的准确率。

关键词: 信息过滤, 粗糙集, 相似度, 同义优化, 回归

Abstract: In order to describe the user’s interest more exactly and provide for him what he needs, a new method to weight and optimize the user’s characterized interest in information filtering is presented. The paper defines the weights of the terms according to the order of the proposed sequence, optimizes the interest with synonyms in rough set theory, calculates the semantic similarity between interest and documents to be filtered, sorts them in descending order of their similarity, and submits them to the user. Experiments show that it has the higher precision.

Key words: Information filtering, Rough sets, Similarity, Synonym optimization, Regress

中图分类号: