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 ›› 2010, Vol. 36 ›› Issue (16): 126-128. doi: 10.3969/j.issn.1000-3428.2010.16.046

• 安全技术 • 上一篇    下一篇

一种改进的隐私保持协同过滤推荐算法

张付志,刘 亭,封素石   

  1. (燕山大学信息科学与工程学院,秦皇岛 066004)
  • 出版日期:2010-08-20 发布日期:2010-08-17
  • 作者简介:张付志(1964-),男,教授,主研方向:智能网络信息处理,网络与信息安全,面向服务计算;刘 亭、封素石,硕士研究生
  • 基金资助:
    国家“973”计划基金资助项目(2005CB32190);河北省自然科学基金资助项目(F2008000877)

Improved Privacy-preserving Collaborative Filtering Recommendation Algorithm

ZHANG Fu-zhi, LIU Ting, FENG Su-shi   

  1. (School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004)
  • Online:2010-08-20 Published:2010-08-17

摘要: 采用随机扰乱技术的协同过滤推荐算法会降低推荐精度,基于此,提出扰乱强度权重的概念及其度量方法,给出一种改进的基于随机扰乱技术的隐私保持协同过滤推荐算法。该算法依据推荐用户的扰乱强度计算相应的扰乱强度权重,相似度的计算综合考虑用户评分相似度和扰乱强度权重两方面因素。实验表明,改进后的算法在不影响隐私保护效果的前提下,提高了推荐精度。

关键词: 协同过滤, 随机扰乱, 隐私保护, 扰乱强度权重, 相似度

Abstract: To solve the accuracy decreasing problem of the recommendation algorithm using Randomized Perturbation Techniques(RPT), the concept of perturbation intensity weight, its measurement and an improved method for similarity calculation are introduced in this paper. An improved privacy preserving collaborative filtering recommendation algorithm is also proposed. The algorithm calculates each recommender’s perturbation intensity weight according to its perturbation intensity. Both users’ rating similarity and perturbation intensity weight are considered when similarity is calculated. Experimental results show that the improved algorithm outperforms the initial one in accuracy without affecting the effectiveness of privacy protection.

Key words: collaborative filtering, randomized perturbation, privacy protection, perturbation intensity weight, similarity

中图分类号: