基于深度距离分解的推荐算法研究

摘要
参考文献
相关文章 (10)

全文: PDF (1736 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对推荐系统中的矩阵分解算法只根据用户和物品的特征向量进行点积运算,无法准确衡量不同用户对物品偏好的弊端,该文提出了一种基于深度距离分解模型的推荐算法。首先,改变传统矩阵分解直接对评分值进行分解的模式,将用户与物品的评分矩阵转化为距离矩阵;然后,将距离矩阵分别按行和按列输入两个深度神经网络进行训练,得到用户和物品的距离特征向量;接下来,用距离特征向量计算用户和物品之间的距离值,通过设计的损失函数使预测距离值与真实距离值的误差达到最小;最后,将用户与物品的预测距离值转化为预测评分。实验结果表明,在不同数据集中,该文提出的基于深度距离分解模型的推荐算法在RMSE和MAE指标上均优于基线推荐算法。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	钱梦薇
	过弋

关键词 ：距离分解, 深度神经网络, 评分预测

Abstract：To deal with the issue that the dot product adopted in matrix factorization can’t accurately measure users’ preference for items, a deep distance factorization model for recommender system is proposed. Firstly, the user-item rating matrix is converted into a distance matrix instead of being directly decomposed. Next, the distance matrix is input into two deep neural networks by row and column, and the distance feature vectors of users and items are obtained. Then, the distance between the user and the item is calculated with the distance feature vectors of users and items, and the error between predicted distance value and real distance value is minimized through the designed loss function. Finally, the ratings are converted from the predicted distance values. Experiments on different datasets show that the proposed algorithm outperforms other algorithms on rating prediction task.

Key words： distance factorization deep neural networks ratings prediction

收稿日期: 2020-06-16

中图分类号:

TP391

基金资助:国家重点研发计划(2018YFC0807105);上海市科学技术委员会科研计划项目(17DZ1101003,18511106602,18DZ2252300)

作者简介: 钱梦薇(1996—),硕士,主要研究领域为文本挖掘、推荐系统。E-mail:ww649960358@foxmail.com
过弋(1975—),通信作者,博士,教授,主要研究领域为文本挖掘、知识发现、商业智能分析。E-mail:guoyi@ecust.edu.cn

链接本文:

http://jcip.cipsc.org.cn/CN/Y2022/V36/I5/41

[1] Alharthi H, Inkpen D, Szpakowicz S. A survey of book recommender systems[J]. Journal of Intelligent Information Systems, 2018, 51 (1): 139-160.
[2] Zhang S, Yao L, Huang C, et al. Metric factorization: recommendation beyond matrix factorization[EB/OL]. http://xxx.itp.ac. cn/pdf/1802.04606v2.[2018-06-04].
[3] Yang X, Guo Y, Liu Y, et al. A survey of collaborative filtering based social recommender systems[J]. Computer Communications, 2014, 41 (5): 1-10.
[4] Aaldering L J, Leker J, Song C H. Recommending untapped M& A opportunities: a combined approach using principal component analysis and collaborative filtering[J]. Expert Systems with Application, 2019, 125 (7): 221-232.
[5] 蒋宗礼,于莉.基于用户特征的协同过滤推荐算法[J].计算机系统应用, 2019, 28 (08): 190-196.
[6] Karabadji N E I, Beldjoudi S, Seridi H, et al. Improving memory-based user collaborative filtering with evolutionary multi-objective optimization[J]. Expert systems with applications, 2018, 98: 153-165.
[7] Zhang L, Liu X, Cao Y, et al. O-recommend: an optimized user-based collaborative filtering recommendation system [C]//Proceedings of IEEE International Conference on Parallel and Distributed Systems: IEEE, 2018: 212-219.
[8] Sebastian S, Ufuk C, Ted D. Efficient incremental cooccurrence analysis for item-based collaborative filtering [C]//Proceedings of the 31st International Conference on Scientific and Statistical Database Management, 2019: 61-72.
[9] Najafabadi M K, Mohame A, Onn C W. An impact of time and item influencer in collaborative filtering recommendations using graph-based model [J]. Information Processing and Management, 2019, 55 (3): 526-540.
[10] Guo T L, Luo J Z, Dong K, et al. Locally differentially private item-based collaborative filtering[J]. Information Sciences, 2019, 502: 229-246.
[11] LUO Y, SUN L, CHEN F, et al. A new recommender system using context clustering based on matrix factorization techniques[J]. Chinese Journal of Electronics, 2016, 25 (2): 334-340.
[12] 刘晴晴, 罗永龙, 汪逸飞,等. 基于SVD填充的混合推荐算法[J]. 计算机科学, 2019, 46 (z1): 468-472.
[13] Shi W C, Wang L J, Qin J W. User embedding for rating prediction in SVD++-based collaborative filtering[J]. Symmetry, 2020, 12 (1): 121.
[14] Zhang S, Liu L X, Chen Z L, et al. Probabilistic matrix factorization with personalized differential privacy[J]. Knowledge-Based Systems, 2019, 183: 104864.
[15] 王运, 倪静. 融合用户偏好和物品相似度的概率矩阵分解推荐算法[J]. 小型微型计算机系统, 2020, 41 (04): 746-751.
[16] Xiao T, Shen H. Neural variational matrix factorization for collaborative filtering in recommendation systems[J]. Applied Intelligence, 2019, 49 (10): 3558-3569.
[17] Chen C, Zhang M, Zhang Y, et al. Efficient neural matrix factorization without sampling for recommendation[J]. ACM Transactions on Information Systems, 2020, 38 (2): 1-28.
[18] 邝神芬, 黄业文, 宋杰, 李洽. 基于深度矩阵分解网络的矩阵填充方法[J]. 计算机科学, 2019, 46 (10): 55-62.
[19] Li P J, Wang Z H, Ren Z C, et al. Neural rating regression with abstractive tips generation for recommendation [C]//Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2017: 345-354.
[20] Suvash S, Aditya K M, Scott S, et al. AutoRec: autoencoders meet collaborative filtering [C]//Proceedings of the 24th International Conference on World Wide Web, 2015: 111-112.