首页 | 本学科首页   官方微博 | 高级检索  
     

融合争议度特征的协同过滤推荐算法
引用本文:张学胜,陈超,张迎峰,俞能海.融合争议度特征的协同过滤推荐算法[J].小型微型计算机系统,2012,33(4):694-699.
作者姓名:张学胜  陈超  张迎峰  俞能海
作者单位:1. 中国科学技术大学多媒体计算与通信教育部-微软重点实验室,合肥,230027
2. 中国科学技术大学图书馆,合肥,230027
基金项目:国家“八六三”高技术研究发展计划项目,国家自然科学基金重点项目,国家重大科技专项,博士学科点专项科研基金
摘    要:基于项目的协同过滤推荐算法在电子商务中有着广泛的引用,该算法的核心是计算项目之间的相似度.传统的计算项目相似度算法仅仅通过项目间共同用户评分值差异来计算,在数据稀疏情况下,项目间共同用户评分值很少,导致此类算法性能严重下降.针对此问题,从项目间的整体评分角度出发,提出争议相似度的概念,争议相似度从项目间评分方差差异的角度衡量项目间相似性.将争议度特征融合到基于项目之间共同用户评分的传统相似度算法中,进而提出了融合项目争议度特征的协同过滤推荐算法,最终缓解了传统算法在稀疏数据情况下相似度计算不准确的问题.实验结果表明该算法在数据稀疏环境下可以明显提升推荐质量.

关 键 词:协同过滤  最近邻居  争议相似度  数据稀疏

Collaborative Filtering Algorithm Combining the Feature of Controversy
ZHANG Xue-sheng , CHEN Chao , ZHAGN Ying-feng , YU Neng-hai.Collaborative Filtering Algorithm Combining the Feature of Controversy[J].Mini-micro Systems,2012,33(4):694-699.
Authors:ZHANG Xue-sheng  CHEN Chao  ZHAGN Ying-feng  YU Neng-hai
Affiliation:1(MOE-MS Key Lab of MCC University of Science and Technology of China,University of Science and Technology of China,Hefei 230027,China) 2(Library of University of Science and Technology of China,Hefei 230027,China)
Abstract:Item-based Collaborative Filtering(CF) algorithm has been widely used in e-commerce.The most critical component of the algorithm is how to measure the similarity between items.Traditional calculations of similarities relied on the scores of the items that two users both rated,which suffers from data sparsity and poor prediction quality problems.In this paper,we consider the whole ratings between items and propose the conception of "Item Controversy Similarity(ICS)",which measures the items′ similarity by calculating the divergence of variance of the rating values between items.Combing the ICS to the traditional similarity calculation algorithm,we propose a new CF algorithm,which could reduce the inaccurate similarity in data sparsity.Empirical studies on dataset MovieLens show that algorithm outperforms other state-of-the-art CF algorithms and it is more robust against data sparsity.
Keywords:collaborative filtering  nearest neighbors  controversy similarity  data sparsity
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号