基于时间衰减的协同过滤推荐算法

doi:10.13229/j.cnki.jdxbgxb201704036

吉林大学学报(工学版) ›› 2017, Vol. 47 ›› Issue (4): 1268-1272.doi: 10.13229/j.cnki.jdxbgxb201704036

基于时间衰减的协同过滤推荐算法

董立岩^{1, 2}, 王越群¹, 贺嘉楠¹, 孙铭会^{1, 2}, 李永丽³

1.吉林大学计算机科学与技术学院,长春 130012;
2.吉林大学符号计算与知识工程教育部重点实验室,长春 130012;
3.东北师范大学信息科学与技术学院,长春 130117

收稿日期:2016-12-30 出版日期:2017-07-20 发布日期:2017-07-20
通讯作者: 孙铭会(1983-),男,讲师,博士.研究方向:人机交互,数据挖掘.E-mail:smh@jlu.edu.cn
作者简介:董立岩(1966-),男,教授,博士生导师.研究方向:数据挖掘.E-mail:dongly@jlu.edu.cn
基金资助:
国家自然科学基金项目(61272209,61300145).

Collaborative filtering recommendation algorithm based on time decay

DONG Li-yan^{1, 2}, WANG Yue-qun¹, HE Jia-nan¹, SUN Ming-hui^{1, 2}, LI Yong-li³

1.College of Computer Science and Technology,Jilin University,Changchun 130012,China;
2.Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education,Jilin University,Changchun 130012;
3.School of Computer Science and Information Technology,Northeast Normal University,Changchun 130117,China

Received:2016-12-30 Online:2017-07-20 Published:2017-07-20

摘要/Abstract

摘要： 针对传统的协同过滤算法在计算相似度时未考虑时间因素的影响,导致推荐结果不准确的问题,本文提出将时间因素融入用户项目评分矩阵中,以解决兴趣衰减的问题。首先将遗忘曲线和记忆周期作为时间因素融入算法之中,将艾宾浩斯遗忘曲线用于指数函数拟合,从而获得时间与兴趣衰减的函数关系,以此用于优化用户项目的评分。并将改进的评分矩阵应用到基于项目的协同过滤推荐算法中进行推荐。在评分中加入记忆周期的影响,让目标用户对待预测的项目评分预测更为准确。实验结果表明,改进后的基于时间衰减协同过滤算法在准确性方面有显著的提高。

关键词: 计算机系统结构, 协同过滤, 推荐算法, 时间衰减曲线, 相似度

Abstract: Since time factor is not considered in calculating score matrix in traditional collaborative filtering, the accuracy for recommendation is limited. In order to solve this problem, we propose a method, which integrates the time factor into the user score matrix. In this method, first the forgetting curve and the memory period are incorporated into the algorithm as the time factor. Then Ebbinghaus forgetting curve is fitted with potential function. This way can get more correct correlation function between time and interest decay. Finally all the results are imported into the user score matrix, and then the matrix is applied in collaborative filtering recommendation algorithm. The factors, which add the memory cycle to the score, make the target user predict the predicted project score more accurately. The experiments show that the results obtained with the proposed method are much more accurate than that of the original one.

Key words: computer systems organization, collaborative filtering(CF), recommendation algorithm, time decay curve, accuracy

中图分类号:

TP393

董立岩, 王越群, 贺嘉楠, 孙铭会, 李永丽. 基于时间衰减的协同过滤推荐算法[J]. 吉林大学学报(工学版), 2017, 47(4): 1268-1272.

DONG Li-yan, WANG Yue-qun, HE Jia-nan, SUN Ming-hui, LI Yong-li. Collaborative filtering recommendation algorithm based on time decay[J]. 吉林大学学报(工学版), 2017, 47(4): 1268-1272.

参考文献

[1] Lyu L, Medo M, Yeung C H, et al. Recommender systems[J]. Physics Reports, 2012, 519(1): 1-49.
[2] 景民昌. 从ACM RecSys'2014国际会议看推荐系统的热点和发展[J]. 现代情报,2015, 35(4):41-45.
Jing Min-chang. Hot topics and trends of recommender system research:areview on ACM RecSys'2014 annual meeting[J]. Journal of Modern Information,2015,35(4):41-45.
[3] 冷亚军,陆青,梁昌勇. 基于结构相似性的协同过滤推荐算法[J]. 小型微型计算机系统,2015,36(10): 2266-2269.
Leng Ya-jun, Lu Qing, Liang Chang-yong. Collaborative filtering recommendation algorithm based on structure similarity[J]. Journal of Chinese Computer Systems,2015, 36(10): 2266-2269.
[4] 汪静, 印鉴, 郑利荣,等. 基于共同评分和相似性权重的协同过滤推荐算法[J]. 计算机科学, 2010, 37(2):99-104.
Wang Jing, Yin Jian, Zheng Li-rong, et al. Collaborative filtering recommendation algorithm based on co-ratings and similarity weight[J].Computer Science,2010,37(2): 99-104.
[5] Choi K, Suh Y. A new similarity function for selecting neighbors for each target item in collaborative filtering[J]. Knowledge-Based Systems, 2013, 37(1):146-153.
[6] 朱思丞, 黄瑛, 孙志锋. 推荐算法时间动态特性研究进展[J]. 工业控制计算,2015,28(8):99-100.
Zhu Si-cheng, Huang Ying, Sun Zhi-feng. Research on progress of time-based dynamic recommender system[J]. Industrial Control Computer, 2015,28(8):99-100.
[7] Seminario C E, Wilson D C. Attacking item-based recommender systems with power items[C]//Proceedings of the 8th ACM Conference on Recommender Systems, New York,2014: 57-64.
[8] Zhang Xiang-liang, Lee T, Pitsilis G. Securing recommender systems against shilling attacks using social-based clustering[J].Journal of Computer Science and Technology, 2013, 28(4): 616-624.
[9] Zhang L, Xu J, Li C. A random-walk based recommendation algorithm considering item categories[J].Neurocomputing,2013,120(10):391-396.
[10] Bobadilla J, Ortega F, Hernando A, et al. Recommender systems survey[J]. Knowledge-Based Systems ,2013, 46: 109-132.
[11] Qian X, Feng H, Zhao G, et al. Personalized recommendation combining user interest and social circle[J]. IEEE transactions on Knowledge and Data Engineering, 2014, 26(7): 1763-1777.
[12] Zeng L, Lin L. An Interactive vocabulary learning system based on word frequency lists and ebbinghaus' curve of forgetting[C]//Digital Media and Digital Content Management(DMDCM), NJ:IEEE Computer Society Press,2011: 313-317.
[13] Wang J, Hu X, Li Z, et al. Learning to recommend questions based on public interest[C]//ACM International Conference on Information and Knowledge Management,NY:ACM Press,2011:2029-2032.
[14] Guo Y, Huang M, Sun L. A collaborative filtering algorithm of selecting neighbors for each target item[C]//Web Information System and Application Conference,NJ:IEEE Computer Society Press,2014:139-143.
[15] Li Z, Lin J. An analysis of the subscription in user-generated content video systems[C]//International Conference on Computer Communications and Networks,Washington:IEEE Computer Society Press,2012:1-7.

相关文章 15

[1]	余宜诚, 胡亮, 迟令, 初剑峰. 一种改进的适用于多服务器架构的匿名认证协议[J]. 吉林大学学报(工学版), 2018, 48(5): 1586-1592.
[2]	桂春, 黄旺星. 基于改进的标签传播算法的网络聚类方法[J]. 吉林大学学报(工学版), 2018, 48(5): 1600-1605.
[3]	王旭, 欧阳继红, 陈桂芬. 基于垂直维序列动态时间规整方法的图相似度度量[J]. 吉林大学学报(工学版), 2018, 48(4): 1199-1205.
[4]	王旭, 欧阳继红, 陈桂芬. 基于多重序列所有公共子序列的启发式算法度量多图的相似度[J]. 吉林大学学报(工学版), 2018, 48(2): 526-532.
[5]	董坚峰, 张玉峰, 戴志强. 改进的基于狄利克雷混合模型的推荐算法[J]. 吉林大学学报(工学版), 2018, 48(2): 596-604.
[6]	赵博, 秦贵和, 赵永哲, 杨文迪. 基于半陷门单向函数的公钥密码[J]. 吉林大学学报(工学版), 2018, 48(1): 259-267.
[7]	曹婧华, 孔繁森, 冉彦中. 面向服务推荐的QoS成列协同排序算法[J]. 吉林大学学报(工学版), 2018, 48(1): 274-280.
[8]	刘磊, 刘利娟, 吴新维, 张鹏. 基于ECPMR的编译器测试方法[J]. 吉林大学学报(工学版), 2017, 47(4): 1262-1267.
[9]	于斌斌, 武欣雨, 初剑峰, 胡亮. 基于群密钥协商的无线传感器网络签名协议[J]. 吉林大学学报(工学版), 2017, 47(3): 924-929.
[10]	邓昌义, 郭锐锋, 张忆文, 王鸿亮. 基于平衡因子的动态偶发任务低功耗调度算法[J]. 吉林大学学报(工学版), 2017, 47(2): 591-600.
[11]	魏晓辉, 刘智亮, 庄园, 李洪亮, 李翔. 支持大规模流数据在线处理的自适应检查点机制[J]. 吉林大学学报(工学版), 2017, 47(1): 199-207.
[12]	郝娉婷, 胡亮, 姜婧妍, 车喜龙. 基于多管理节点的乐观锁协议[J]. 吉林大学学报(工学版), 2017, 47(1): 227-234.
[13]	王贵参, 黄岚, 王岩, 宋立明, 欧歌. 引入极值非相邻连接的连接聚类方法[J]. 吉林大学学报(工学版), 2016, 46(5): 1616-1621.
[14]	魏晓辉, 李翔, 李洪亮, 李聪, 庄园, 于洪梅. 支持大规模流数据处理的弹性在线MapReduce模型及拓扑协议[J]. 吉林大学学报(工学版), 2016, 46(4): 1222-1231.
[15]	车翔玖, 梁森. 一种基于大顶堆的SPIHT改进算法[J]. 吉林大学学报(工学版), 2016, 46(3): 865-869.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于时间衰减的协同过滤推荐算法

Collaborative filtering recommendation algorithm based on time decay

RICH HTML

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 10