Page 253 - 《软件学报》2021年第12期

软件学报 ISSN 1000-9825, CODEN RUXUEW E-mail: jos@iscas.ac.cn
Journal of Software,2021,32(12):3917−3928 [doi: 10.13328/j.cnki.jos.006141] http://www.jos.org.cn
©中国科学院软件研究所版权所有. Tel: +86-10-62562563

∗
深度矩阵分解推荐算法

1
1
1
田 震 , 潘腊梅 , 尹 朴 , 王 睿 1,2
1
(北京科技大学 计算机与通信工程学院,北京 100083)
2 (北京科技大学 顺德研究生院,广东 佛山 528300)
通讯作者: 王睿, E-mail: wangrui@ustb.edu.cn

摘 要: 协同过滤推荐算法中的矩阵分解因其简单、易于实现,得到了广泛的应用.但是矩阵分解通过简单的线
性内积建模用户和物品之间的非线性交互关系,限制了模型的表达能力.为此,He 等人提出了广义矩阵分解模型,通
过非线性激活函数和连接权重,将矩阵分解推广到广义矩阵分解,为模型赋予建模用户和物品间的二阶非线性交互
关系的能力.但是广义矩阵分解模型是一个浅层模型,并不能很好地建模用户和物品间高阶交互关系,一定程度上可
能会影响模型性能.受广义矩阵分解模型启发,提出了深度矩阵分解模型(deep matrix factorization,简称 DMF),在广
义矩阵分解模型的基础上引入隐藏层,利用深层神经网络来学习用户和物品间高阶交互关系.深度矩阵分解模型不
仅解决了简单内积的线性问题,同时还能够建模用户和物品间的高阶交互,具有很好的表达能力.此外,在 MovieLens
和 Anime 两个数据集上进行了大量丰富的对比实验,验证了模型的可行性和有效性;同时,通过实验确定了模型的最
优参数.
关键词: 协同过滤;线性内积;广义矩阵分解;隐藏层;高阶交互
中图法分类号: TP311

中文引用格式: 田震,潘腊梅,尹朴,王睿.深度矩阵分解推荐算法.软件学报,2021,32(12):3917−3928. http://www.jos.org.cn/
1000-9825/6141.htm
英文引用格式: Tian Z, Pan LM, Yin P, Wang R. Deep matrix factorization recommendation algorithm. Ruan Jian Xue Bao/
Journal of Software, 2021,32(12):3917−3928 (in Chinese). http://www.jos.org.cn/1000-9825/6141.htm
Deep Matrix Factorization Recommendation Algorithm

1
1
1
TIAN Zhen , PAN La-Mei , YIN Pu , WANG Rui 1,2
1
(School of Computer & Communication Engineering, University of Science and Technology Beijing, Beijing 100083, China)
2
(Shunde Graduate School, University of Science and Technology Beijing, Foshan 528300, China)
Abstract: Matrix factorization in collaborative filtering recommendation algorithms is widely used because of its simplicity and facility
of implementation, but matrix factorization utilizes a simple linear inner product to model the non-linear interaction between the user and
the item, which limits the model's expressive power. He et al. proposed a generalized matrix factorization model, which extended the
matrix factorization to the generalized matrix factorization through a non-linear activation function and connection weights, and gave the
model the ability to model second-order non-linear interactions between users and items. Nevertheless, the generalized matrix
factorization model is a shallow model and does not model the high-order interaction between users and items, which may affect the
performance of the model to a certain extent. Inspired by the generalized matrix factorization model, this study proposes a deep matrix
factorization model, abbreviated as DMF. Based on the generalized matrix factorization model, a hidden layer is introduced, and a deep
neural network is used to learn the higher-order interaction between users and items. The deep matrix factorization model, which has a
good expression ability, not only solves the linear problem of simple inner product, but also models high-order interactions between users

∗ 基金项目: 国家自然科学基金(62173158, 61803391); 北京科技大学顺德研究生院科技创新专项资金(BK19CF010, BK20BF012)
Foundation item: National Natural Science Foundation of China (62173158, 61803391); Scientific and Technological Innovation
Foundation of Shunde Graduate School, USTB (BK19CF010, BK20BF012)
收稿时间: 2020-03-31; 修改时间: 2020-06-11; 采用时间: 2020-08-28