GNN-Based Matrix Factorization Recommendation Algorithm

doi:10.3778/j.issn.1002-8331.2009-0013

Abstract

Abstract:

Compared with collaborative filtering, matrix factorization has better scalability and flexibility, but it is also troubled by data sparseness and cold start. Aiming at the above problems, a recommendation algorithm GNN_MF combining Graph Neural Network（GNN） and Probabilistic Matrix Factorization（PMF） is proposed. The algorithm uses GNN to model social network graphs and user item graphs, connects the two graphs internally, and learns the feature vector of the target user in the social space and item space. Then through Multi-Layer Perceptron（MLP）, the two feature vectors are connected in series to extract the user’s potential feature vector. Finally, it is integrated on the probability matrix factorization model to generate prediction scores. A large number of experiments on real data sets Epinions and Ciao show that the root mean square error and average absolute error of the GNN_MF algorithm are reduced by 2.91%, 3.10% and 4.83%, 3.84% respectively compared with traditional PMF. The effectiveness and feasibility of the GNN_MF algorithm in the recommendation system are verified.

Key words: probability matrix decomposition, Graph Neural Network（GNN）, recommendation algorithm, social network

摘要：

相较于协同过滤，矩阵分解有着更好的拓展性和灵活性，但同样受到数据稀疏和冷启动的困扰。针对上述问题，提出一种融合GNN和PMF的推荐算法GNN_MF。该算法通过神经网络对社交网络图以及用户项目图进行建模，将两个图内在的联系起来，学习目标用户在社会空间以及项目空间上的特征向量。通过MLP将两个特征向量串联提取用户的潜在特征向量，集成在概率矩阵分解模型上，产生预测评分。在真实数据集Epinions、Ciao上的大量实验表明，GNN_MF算法的均方根误差和平均绝对误差较传统PMF分别降低了2.91%、3.10%和4.83%、3.84%。验证了GNN_MF算法在推荐系统中的有效性以及可行性。

关键词: 概率矩阵分解, 图神经网络, 推荐算法, 社交网络

WANG Yingbo, SUN Yongdi. GNN-Based Matrix Factorization Recommendation Algorithm[J]. Computer Engineering and Applications, 2021, 57(19): 129-134.

王英博，孙永荻. 基于GNN的矩阵分解推荐算法[J]. 计算机工程与应用, 2021, 57(19): 129-134.

References

[1] 李同欢，唐雁，刘冰.基于深度学习的多交互混合推荐模型[J].计算机工程与应用，2019，55（1）：135-141.
LI T H，TANG Y，LIU B.Multi-interaction hybrid recommendation model based on deep learning[J].Computer Engineering and Applications，2019，55（1）：135-141.
[2] SARWAR B M，KARYPIS G，KONSTAN J A，et al.Item-based collaborative filtering recommendation algorithms[C]//Proceedings of the 10th International Conference on World Wide Web，2001：285-295.
[3] 马宏伟，张光卫，李鹏.协同过滤推荐算法综述[J].小型微型计算机系统，2009，30（7）：1282-1288.
MA H W，ZANG G W，LI P.Summary of collaborative filtering recommendation algorithm[J].Journal of Computer Systems，2009，30（7）：1282-1288.
[4] HE X，ZHANG H，KAN M Y，et al.Fast matrix factorization for online recommendation with implicit feedback[C]//Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval，2016：549-558.
[5] KOREN Y.Factorization meets the neighborhood：Amultifaceted collaborative filtering model[C]//Proceedings of the 14th ACM SIGKDD Conference on Knowledge Discovery and Data Mining，2008：426-434.
[6] MA H，YANG H，LYU M R，et al.Sorec：Social recommendation using probabilistic matrix factorization[C]//Proceedings of the 17th ACM Conference on Information and Knowledge Management，2008：931-940.
[7] MNIH A，SALAKHUTDINOV R R.Probabilistic matrix factorization[C]//Proceedings of Advances in Neural Information Processing Systems，2008：1257-1264.
[8] 周洋，陈家琪.基于栈式降噪自编码器的协同过滤算法[J].计算机应用研究，2017，34（8）：2336-2339.
ZHOU Y，CHEN J Q.Collaborative filtering algorithm based on stacked noise reduction autoencoder[J].Application Research of Computers，2017，34（8）：2336-2339.
[9] 曾安，赵恢真.融合了LSTM和PMF的推荐算法[J].计算机工程与应用，2020，56（19）：68-75.
ZENG A，ZHAO H Z.Recommendation algorithm combining LSTM and PMF[J].Computer Engineering and Applications，2020，56（19）：68-75.
[10] HE X，LIAO L，ZHANG H，et al.Neural collaborative filtering[C]//Proceedings of the 26th International ACM SIGIR Conference on Research and Development in Information Retrieval，2017：355-364.
[11] FAN W，LI Q，CHENG M，et al.Deep modeling of social relations for recommendation[C]//Proceedings of the 2018 National Conference on Artificial Intelligence，2018：8075-8076.
[12] DENG S，HUANG L，XU G，et al.On deep learning for trust-aware recommendations in social networks[J].IEEE Transactions on Neural Networks & Learning Systems，2017，28（5）：1164-1177.
[13] DEFFERRARD M，BRESSON X，VANDERGHEYNST P.Convolutional neural networks on graphs with fast localized spectral filtering[C]//Advances in Neural Information Processing Systems，2016：3844-3852.
[14] DERR T，MA Y，TANG J.Signed graph convolutional networks[C]//2018 IEEE International Conference on Data Mining（ICDM），2018：929-934.
[15] BERG R V D，KIPF T N，WELLING M.Graph convolutional matrix completion[J].arXiv：1706.02263.2017.
[16] MONTI F，BRONSTEIN M M，BRESSON X.Geometric matrix completion with recurrent multi-graph neural networks[C]//Advances in Neural Information Processing Systems，2017：3697-3707.
[17] YING R，HE R，CHEN K，et al.Graph convolutional neural networks for Web-scale recommender systems[C]//Proceedings of ACM International Conference on Knowledge Discovery and Data Mining，2018.
[18] 李娴，张泽华，赵霞，等.TWD-GNN：基于三支决策的图神经网络推荐方法[J].计算机工程与应用，2020，56（12）：156-162.
LI X，ZHANG Z H，ZHAO X，et al.TWD-GNN：A graph neural network recommendation method based on three-branch decision[J].Computer Engineering and Applications，2020，56（12）：156-162.
[19] 田保军，刘爽，房建东.融合主题信息和卷积神经网络的混合推荐算法[J].计算机应用，2020，40（7）：1901-1907.
TIAN B J，LIU S，FANG J D.Hybrid recommendation algorithm combining topic information and convolutional neural network[J].Computer Applications，2020，40（7）：1901-1907.
[20] FAN W，MA Y，LI Q，et al.Graph neural networks for social recommendation[C]//Proceedings of the World Wide Web Conference，2019：417-426.