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 ›› 2011, Vol. 37 ›› Issue (15): 271-273. doi: 10.3969/j.issn.1000-3428.2011.15.088

• 开发研究与设计技术 • 上一篇    下一篇

基于用户行为的盲人图书推荐方法

方 彬1,胡 侠2,王 灿1   

  1. (1. 浙江大学计算机科学与技术学院,杭州 310027;2. 杭州市科技信息研究院,杭州 310001)
  • 收稿日期:2010-12-27 出版日期:2011-08-05 发布日期:2011-08-05
  • 作者简介:方 彬(1985-),男,硕士,主研方向:信息检索;胡 侠,助理研究员、硕士;王 灿,讲师
  • 基金资助:
    国家科技支撑计划基金资助项目(2008BAH26B00)

Book Recommendation Method for the Blind Based on User Behavior

FANG Bin  1, HU Xia  2, WANG Can  1   

  1. (1. College of Computer Science & Technology, Zhejiang University, Hangzhou 310027, China; 2. Hangzhou Science & Technology Information Institute, Hangzhou 310001, China)
  • Received:2010-12-27 Online:2011-08-05 Published:2011-08-05

摘要: 为帮助盲人更有效便捷地通过网络获取所需书籍资源,提出一种基于用户行为的图书推荐方法。该方法在考虑用户兴趣随时间变化的基础上,通过分析用户日志,建模表达访问书籍、访问书籍类别、使用书签3类用户行为,实现盲人用户的图书推荐,同时缓解盲人数字图书馆中的用户数据稀疏问题。实验结果证明,该方法推荐效果良好。

关键词: 用户行为, 协同过滤, 盲人图书推荐系统, 推荐方法, 用户数据稀疏

Abstract: In order to help the blind access to the resources of e-books more effectively, this paper proposes a book recommendation method based on users’ behavior. By analyzing users’ logs and considerations of the time factor effect on users’ interests, it proposes a rational behavior model given data of user access to books, book categories and the usage of bookmarks to generate the rational book recommendation for the blind. It helps relief the problem which is caused by user data sparseness in blind digital library. Experimental results show that the algorithm can provide good recommendation results.

Key words: user behavior, collaborative filtering, book recommender system for the blind, recommendation method, user data sparseness

中图分类号: