融合知识图谱和用户行为信息的个性化推荐算法研究
Research on Personalized Recommendation Algorithm Based on Knowledge Graph and User Behavior Information
摘要: 针对传统协同过滤存在的稀疏性和冷启动问题，通常使用深度神经网络(DNN)构建融合知识图谱和推荐系统的推荐任务。但目前的方法未曾考虑特征间的低阶线性关系，虽可加入因子分解机(FM)，但不同的特征对模型的贡献不同，FM可能会因所有特征交互设置相同的权重而受到阻碍；DNN解决知识图谱这种具有不规则、可扩展、多重结构特性的数据结构不具普适性。针对以上问题，提出MKAFG模型，推荐部分加入具有注意力机制的FM，通过注意力网络区分不同特征交互的重要性，使FM提取到对目标预测起到重要作用的一阶、二阶线性交互特征。知识嵌入部分使用图卷积神经网络(GCN)，提高推荐系统推荐效果。实验结果表明，MKAFG在MovieLens-1M数据集上的推荐效果优于主流推荐模型。
Abstract: To solve the sparsity and cold start problems of traditional collaborative filtering, deep neural net-work (DNN) is usually used to construct the recommendation task of fusion knowledge map and recommendation system. However, the current method does not consider the low-order linear rela-tionship between features. Although factor decomposition machine (FM) can be added, different features have different contributions to the model, and FM may be hindered because all features set the same weight interactively. DNN solves the problem that knowledge map, a data structure with irregular, extensible and multiple structural characteristics, is not universal. To solve the above problems, the MKAFG model is proposed, and FM with attention mechanism is recommended. The importance of interaction between different features is distinguished by attention network, so that FM can extract first-order and second-order linear interactive features which play an important role in target prediction. In the part of knowledge embedding, graph convolution neural network (GCN) is used to improve the recommendation effect of recommendation system. Experimental results show that MKAFG's recommendation effect on MovieLens-1M dataset is better than that of main-stream recommendation models.
文章引用：程静文, 王全民. 融合知识图谱和用户行为信息的个性化推荐算法研究[J]. 计算机科学与应用, 2021, 11(4): 948-961. https://doi.org/10.12677/CSA.2021.114098

参考文献

[1] Wang, H., Wang, J., Wang, J., et al. (2017) GraphGAN: Graph Representation Learning with Generative Adversarial Nets. IEEE Transactions on Knowledge and Data Engineering.
[2] Zhang, F.Z., Yuan, N.J., Lian, D.F., et al. (2016) Collaborative Knowledge Base Embedding for Recommender Systems. In: Proceedings of the 22nd ACM SIGKDD In-ternational Conference on Knowledge Discovery and Data Mining, ACM, New York, 353-362. [Google Scholar] [CrossRef] 
[3] Bahdanau, D., Cho, K. and Bengio, Y. (2016) Neural Machine Translation by Jointly Learning to Align and Translate. Proceedings of the 3rd International Conference on Learning Representations, Banff, September 2014, 353-362.
[4] Wang, H.W., Zhang, F.Z., Xie, X., et al. (2018) DKN: Deep Knowledge-Aware Network for News Recommendation. Proceedings of the 2018 World Wide Web Conference on World Wide Web, Lyon, 23-27 April 2018, 1835-1844. [Google Scholar] [CrossRef] 
[5] Wang, H.W., Zhang, F.Z., Wang, J.L., et al. (2018) RippleNet: Propagating User Preferences on the Knowledge Graph for Recommender Systems. In: Proceedings of the 27th ACM International Conference on Information and Knowledge Management, ACM Press, Torino, 417-426. [Google Scholar] [CrossRef] 
[6] Wang, H.W., Zhang, F.Z., Zhao, M., et al. (2019) Multi-Task Fea-ture Learning for Knowledge Graph Enhanced Recommendation. Proceedings of the 2019 World Wide Web Conference on World Wide Web (WWW 2019), New York, May 2019, 2000-2010. [Google Scholar] [CrossRef] 
[7] Xiao, J., Ye, H. and He, X.N. (2017) Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks. [Google Scholar] [CrossRef] 
[8] Kipf, T.N. and Welling, M. (2017) Semi-Supervised Classification with Graph Convolutional Networks. International Conference on Learning Representations, Toulon, 24-26 April 2017, 14.
https://arxiv.org/abs/1609.02907
[9] Rendle, S. (2010) Factorization Machines. Proceedings of the 10th IEEE In-ternational Conference on Data Mining (ICDM), Sydney, 13-17 December 2010, 995-1000. [Google Scholar] [CrossRef] 
[10] Rendle, S. (2012) Factorization Machines with LibFM. ACM Trans-actions on Intelligent Systems and Technology, 3, 57:1-57:22. [Google Scholar] [CrossRef] 
[11] Pal, S.K. and Mitra, S. (1992) Multilayer Perceptron, Fuzzy Sets, and Classification. IEEE Transactions on Neural Networks, 3, 683-697. [Google Scholar] [CrossRef] [PubMed]
[12] Cheng, H.T., Koc, L., Harmsen, J., et al. (2016) Wide & Deep Learning for Recommender Systems. Workshop on Deep Learning for Recommender Systems, Boston, 15 September 2016, 7-10. [Google Scholar] [CrossRef] 
[13] He, X.N. and Chua, T.S. (2017) Neural Factorization Machines for Sparse Predictive Analytics. Proceedings of the 40th International ACM SIGIR Conference on Research and Develop-ment in Information Retrieval (SIGIR), Tokyo, 7-11 August 2017, 355-364.
[14] Guo, H.F., Tang, R.M., Ye, Y.M., et al. (2017) DeepFM: A Factorization-Machine Based Neural Network for CTR Prediction. Proceedings of the 26th In-ternational Joint Conference on Artificial Intelligence (IJCAI), Melbourne, 19-25 August 2017, 1725-1731. [Google Scholar] [CrossRef] 
[15] Zhou, G., Song, C., Zhu, X., et al. (2017) Deep Interest Network for Click-Through Rate Prediction. Proceedings of the 40th ACM SIGKDD International Conference on Knowledge Dis-covery & Data Mining, June 2017, 1058-1068.
[16] 刘知远, 孙茂松, 林衍凯, 等. 知识表示学习研究进展[J]. 计算机研究与发展, 2016, 53(2): 247-261.
[17] Bordes, A., Weston, J., Collobert, R., et al. (2011) Learning Structured Embeddings of Knowledge Bases. National Conference on Artificial Intelligence, Barcelona, 16-22 July 2011, 301-306.
[18] Bordes, A., Usunier, N., Garciaduran, A., et al. (2013) Translating Embeddings for Modeling Mul-ti-Relational Data. Proceedings of the 26th International Conference on Neural Information Processing Systems, Volume 2, 2787-2795.
[19] Wang, Z., Zhang, J., Feng, J., et al. (2014) Knowledge Graph Embedding by Translating on Hyper-planes. National Conference on Artificial Intelligence, Québec, 27-31 July 2014, 1112-1119.
[20] Lin, Y., Liu, Z., Sun, M., et al. (2015) Learning Entity and Relation Embedding for Knowledge Graph Completion. In: Proceedings of AAAI, AAAI, Menlo Park, 2181-2187.
[21] Yang, S., Korayem, M., Aljadda, K., et al. (2017) Combining Content-Based and Collaborative Filtering for Job Recommendation System: A Cost-Sensitive Statistical Relational Learning Approach. Knowledge-Based Systems, 136, 37-45. [Google Scholar] [CrossRef]