计算机应用 ›› 2013, Vol. 33 ›› Issue (03): 834-837.DOI: 10.3724/SP.J.1087.2013.00834

• 先进计算 • 上一篇    下一篇

基于填充和相似性信任因子的协同过滤推荐算法

郝立燕*,王靖   

  1. 华侨大学 计算机科学与技术学院,福建 厦门 361000
  • 收稿日期:2012-09-14 修回日期:2012-10-03 出版日期:2013-03-01 发布日期:2013-03-01
  • 通讯作者: 郝立燕
  • 作者简介:郝立燕(1988-),男,山东临沂人,硕士研究生,主要研究方向:数据挖掘、推荐系统; 王靖(1981-),男,福建泉州人,副教授,博士,主要研究方向:模式识别、推荐系统。
  • 基金资助:

    国家自然科学基金资助项目(10901062); 福建省高等学校杰出青年科研人才培育计划项目(11FJPY01)。

Collaborative filtering recommendation algorithm based on filling and similarity confidence factor

HAO Liyan*, WANG Jing   

  1. College of Computer Science and Technology, Huaqiao University, Xiamen Fujian 361000, China
  • Received:2012-09-14 Revised:2012-10-03 Online:2013-03-01 Published:2013-03-01
  • Contact: Li-Yan HAO
  • Supported by:

    the National Natural Science Foundation of China

摘要: 为了提高推荐系统在数据稀疏情况下的推荐质量,提出一种改进的协同过滤算法。该方法使用一种数据挖掘算法对稀疏评分矩阵进行填充; 在完整的填充矩阵上计算用户相似性,并引入相似性信任因子; 最终做出推荐预测。典型数据集上的对比实验结果表明,即使在评分数据极为稀疏的情况下,该算法仍能取得较好的结果。

关键词: 推荐, 协同过滤, 稀疏性, 矩阵填充, 信任因子

Abstract: In order to improve the recommendation quality of recommendation system when the data are sparse, an improved collaborative filtering algorithm was proposed. Using a data mining algorithm, the sparse rating matrix was filled firstly. Afterwards user-similarities and their confidence factors were calculated using the complete filling matrix. Ultimately, the recommendation forecast was made. Comparative experiments on typical dataset show that the algorithm is able to achieve better results even with extremely sparse data.

Key words: recommend, collaborative filtering, sparsity, matrix completion, confidence factor

中图分类号: