引入时间效应的SVD++线性回归推荐算法

doi:10.19678/j.issn.1000-3428.0053427

摘要/Abstract

摘要： 针对传统协同过滤算法中的数据稀疏问题，在SVD++算法和线性回归模型的基础上引入时间效应属性，提出一种推荐算法timeSVD++LR。采用SVD++算法将用户和项目信息与隐式反馈信息相融合映射到隐语义空间，将用户和项目之间的交互作用建模为该空间中的内积。通过描述用户和物品在各因子上的特征来解释评分值，在此基础上对时间效应建模，进一步提高预测结果的准确度。根据预测评分矩阵构造特征向量，将原始训练数据作为线性回归模型的输入，采用梯度下降算法优化最终代价函数，生成使得代价函数值最小的参数向量，同时将特征向量和参数向量代入预测模型求解预测评分。在MovieLens数据集上的实验结果表明，与RSVD、SVD++和timeSVD++算法相比，该算法的平均绝对误差和均方根误差均较低，其推荐准确性较高。

关键词: SVD++模型, 时间效应, 特征向量, 线性回归, 推荐算法

Abstract: Aiming at the problem of data sparsity in traditional collaborative filtering algorithm,this paper proposes a recommended algorithm named timeSVD++ LR,which introduces the time effect attribute on the basis of the SVD++ algorithm and linear regression model.The SVD++ algorithm is used to map the user and item information fusion implicit feedback information to the implicit semantic space,and the user-item interaction is modeled as the inner product of the space.The scoring value is explained by describing the characteristics of users and items on various factors,and then the time effect is modeled to further improve the accuracy of the prediction results.At the same time,the eigenvector is constructed according to the prediction scoring matrix.The original training data is used as the input of the linear regression model,and the final cost function is optimized by gradient descent algorithm to generate the parameter vector that minimizes the value of the cost function.The eigenvector and parameter vector are brought into the prediction model to solve the prediction score.Experimental results on the MovieLens dataset show that,compared with the RSVD,SVD++ and timeSVD++ algorithms,the average absolute error and root mean square error of the proposed algorithm are lower,and its recommendation accuracy is higher.

Key words: SVD++ model, time effect, eigenvector, linear regression, recommendation algorithm

中图分类号:

TP312

苏庆, 章静芳, 李小妹. 引入时间效应的SVD++线性回归推荐算法[J]. 计算机工程, 2020, 46(2): 65-71.

SU Qing, ZHANG Jingfang, LI Xiaomei. SVD++ Linear Regression Recommendation Algorithm with Time Effect[J]. Computer Engineering, 2020, 46(2): 65-71.

https://www.ecice06.com/CN/Y2020/V46/I2/65

图/表 5

20200218141221

20200218141230

20200218141234

20200218141239

20200218141243

参考文献

[1] REN Lei.Research on key technologies of recommendation system[D].Shanghai:East China Normal University,2012.(in Chinese) 任磊.推荐系统关键技术研究[D].上海:华东师范大学,2012.
[2] ZHANG Kaihan,LIANG Jiye,ZHAO Xingwang,et al.A collaborative filtering recommendation algorithm based on information of community experts[J].Journal of Computer Research and Development,2018,55(5):968-976.(in Chinese) 张凯涵,梁吉业,赵兴旺,等.一种基于社区专家信息的协同过滤推荐算法[J].计算机研究与发展,2018,55(5):968-976.
[3] SANTOSH K,XIA N,GEORGE K.FISM:factored item similarity models for top-n recommender systems [C]//Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York,USA:ACM Press,2013:659-667.
[4] LIANG D,ALTOSAAR J,CHARLIN L,et al.Factorization meets the item embedding:regularizing matrix factorization with item co-occurrence[C]//Proceedings of the 10th ACM Conference on Recommender Systems.New York,USA:ACM Press,2016:59-66.
[5] WANG Yuxiang,XU Huan.Stability of matrix factorization for collaborative filtering[C]//Proceedings of the 29th International Conference on Machine Learning.New York,USA:ACM Press,2012:417-424.
[6] SZWABE A,CIESIELCZYK M,JANASIEWICZ T.Semantically enhanced collaborative filtering based on RSVD[C]//Proceedings of International Conference on Computational Collective Intelligence.Berlin,Germany:Springer,2011:10-19.
[7] KOREN Y.Factorization meets the neighborhood:a multifaceted collaborative filtering model[C]//Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York,USA:ACM Press,2008:426-434.
[8] GUO G,ZHANG J,YORKE-SMITH N.TrustSVD:collaborative filtering with both the explicit and implicit influence of user trust and of item ratings[C]//Proceedings of the 29th AAAI Conference on Artificial Intelligence.Palo Alto,USA:AAAI Press,2015:123-129.
[9] WU Bin,LOU Zhengzheng,YE Yangdong.Co-regularized matrix factorization recommendation algorithm[J].Journal of Software,2018,29(9):2681-2696.(in Chinese) 吴宾,娄铮铮,叶阳东.联合正则化的矩阵分解推荐算法[J].软件学报,2018,29(9):2681-2696.
[10] LIANG Xiang,QING Yang.Time-dependent models in collaborative filtering based recommender system[C]//Proceedings of 2009 IEEE/WIC/ACM Intermational Conference on Web Intelligence and Intelligent Agent Technologies.Washington D.C.,USA:IEEE Presss,2009:450-457.
[11] SONG Xin,WANG Cuirong.Linear regression based distributed data gathering optimization strategy for wireless sensor networks[J].Chinese Journal of Computers,2012,35(3):568-580.(in Chinese) 宋欣,王翠荣.基于线性回归的无线传感器网络分布式数据采集优化策略[J].计算机学报,2012,35(3):568-580.
[12] CHEN Zhenhong,LAN Yanyan,GUO Jiafeng,et al.Distributed stochastic gradient descent with discriminative aggregating[J].Chinese Journal of Computers,2015,38(10):2054-2063.(in Chinese) 陈振宏,兰艳艳,郭嘉丰,等基于差异合并的分布式随机梯度下降算法[J].计算机学报,2015,38(10):2054-2063.
[13] NIE Peng.A SVM-based multi-dimension factor decision-making model framework[C]//Proceedings of 2016 International Conference on Applied Mechanics,Electronics and Mechatronics Engineering.[S.l.]:WikiCFP,2016:1-6.
[14] ZHU Aiyun,REN Xiaojun.Recommendation model regularized with user trust and ratings bias[J].Modern Computer,2018,615(15):30-35.(in Chinese) 朱爱云,任晓军.基于用户信任和评分偏置的正则化推荐模型[J].现代计算机,2018,615(15):30-35.
[15] RICCI F,ROKACH L,SHAPIRA B,et al.Recom-mender systems handbook[M].Berlin,Germany:Springer,2010.
[16] ZHANG Fuzhi,CHANG Junfeng,ZHOU Quanqiang.Context-aware recommendation algorithm based on fuzzy C-means clustering[J].Journal of Computer Research and Development,2013,50(10):2185-2194.(in Chinese) 张付志,常俊风,周全强.基于模糊C均值聚类的环境感知推荐算法[J].计算机研究与发展,2013,50(10):2185-2194.
[17] ZHU Yuxiao,LYU Linyuan.Evaluation metrics for recommender systems[J].Journal of University of Electronic Science and Technology of China,2012,41(2):163-175.(in Chinese) 朱郁筱,吕琳媛.推荐系统评价指标综述[J].电子科技大学学报,2012,41(2):163-175.
[18] JIANG Wei.Research on some key technologies of recommendation system[D].Chengdu:University of Electronic Science and Technology,2018.(in Chinese) 蒋伟.推荐系统若干关键技术研究[D].成都:电子科技大学,2018.

选择文件类型/文献管理软件名称

选择包含的内容