异质信息融合网络嵌入的注意力偏好推荐方法

doi:10.3778/j.issn.1002-8331.2010-0017

摘要/Abstract

摘要：

基于异质信息网络的推荐方法已成为当前数据挖掘领域的研究热点。但传统基于异质信息网络的推荐方法多存在可解释性缺失和稀疏不一致性问题，导致无法充分挖掘用户潜在的偏好特征，且有效地进行特征融合。因此，提出了一种在异质信息网络中融合网络嵌入的注意力偏好推荐方法（MFFHINE）；利用对称元路径在刻画对象间语义关系上的优势，在对称元路径上随机游走进行网络嵌入来学习用户偏好特征。采用基于注意力机制的偏好权重融合策略将学习到的各个偏好特征有效融合，并将其集成到矩阵分解模型中。通过联合优化矩阵分解模型和融合函数，以进行最终的评分预测任务。在Douban和Yelp真实大规模数据集上对提出的算法进行实验分析。通过对各基准算法进行横向性能比较，在训练集比例、元路径设置、潜在因子维度等方面进行纵向比较。实验结果表明，MFFHINE性能提升显著。

关键词: 网络嵌入, 异质融合, 推荐系统, 注意力机制

Abstract:

Heterogeneous information network based recommendation methods have been a research focus in current data mining field. However, most of the traditional HIN-based recommendation methods exist sparse inconsistency and lack of interpretability, which cannot adequately explore the user’s potential preference features and effectively integrate features. Therefore, the paper proposes an attention preference recommendation method with Fusing Heterogeneous Information Network Embedding（named as MFFHINE）. Symmetric meta path is apt at characterizing the semantic relationships between objects, hence the network embedding method based on random walks on each symmetric meta path is designed to learn preference features of users. Then the preference weights fusion strategy based on attention mechanism is considered to fuse these learned preference features and then integrate them into the extended matrix factorization model. In this way, the matrix factorization model and the fusion function are jointly optimized to perform the score prediction task. Finally, compared with other benchmark algorithms on Douban and Yelp real large-scale datasets, the algorithm experiments analyze on the training set ratio, meta path settings, latent factor dimensions. Experimental results show that the proposed MFFHINE has been improved more significantly than the state of art algorithms.

Key words: network embedding, heterogeneous fusion, recommendation system, attention mechanism

张杰，张月琴，张泽华，刘志鑫，雷祥. 异质信息融合网络嵌入的注意力偏好推荐方法[J]. 计算机工程与应用, 2021, 57(21): 123-131.

ZHANG Jie, ZHANG Yueqin, ZHANG Zehua, LIU Zhixin, LEI Xiang. Attention Preference Recommendation Methods with Fusing Network Embedding in Heterogeneous Information[J]. Computer Engineering and Applications, 2021, 57(21): 123-131.

参考文献

[1] ZHANG S，YAO L N，SUN A X，et al.Deep learning based recommender system：a survey and new perspectives[J].ACM Computing Surveys（CSUR），2019，52（1）：1-38.
[2] LU J，WU D S，MAO M S，et al.Recommender system application developments：a survey[J].Decision Support Systems，2015，74：12-32.
[3] 王卫红，曾英杰.基于聚类和用户偏好的协同过滤推荐算法[J].计算机工程与应用，2020，56（3）：68-73.
WANG W H，ZENG Y J.Collaborative filtering recommendation algorithm based on clustering and user preference[J].Computer Engineering and Applications，2020，56（3）：68-73.
[4] ZHENG L，LU C T，JIANG F，et al.Spectral collaborative filtering[C]//Proceedings of the 12th ACM Conference on Recommender Systems，2018：311-319.
[5] WANG C，LIU Q，WU R Z，et al.Confidence-aware matrix factorization for recommender systems[C]//The Thirty-Second AAAI Conference on Artificial Intelligence，2018：434-442.
[6] ABDI M H，OKEYO G，MWANGI R W.Matrix factorization techniques for context-aware collaborative filtering recommender systems：a survey[J].Computer and Information Science，2018，11（2）.
[7] GAO L，YANG H，WU J，et al.Recommendation with multi-source heterogeneous information[C]//Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence，2018：3378-3384.
[8] 吴宾，娄铮铮，叶阳东.一种面向多源异构数据的协同过滤推荐算法[J].计算机研究与发展，2019，56（5）：1034-1047.
WU B，LOU Z Z，YE Y D.A collaborative filtering recommendation algorithm for multi-source heterogeneous data[J].Journal of Computer Research and Development，2019，56（5）：1034-1047.
[9] HU L，WANG Y，XIE Z Z，et al.Semantic preference-based personalized recommendation on heterogeneous information network[J].IEEE Access，2017，5：19773-19781.
[10] GAO Y，RAN L X.Collaborative filtering recommendation algorithm for heterogeneous data mining in the internet of things[J].IEEE Access，2019，7：123583-123591.
[11] ZHAO J，ZHOU Z，GUAN Z Y，et al.IntentGC：ascalable graph convolution framework fusing heterogeneous information for recommendation[C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining，2019：2347-2357.
[12] SUN Y Z，HAN J W，YAN X F，et al.Pathsim：meta path-based top-k similarity search in heterogeneous information networks[J].Proceedings of the VLDB Endowment，2011，4（11）：992-1003.
[13] SHI C，HU B B，ZHAO W X，et al.Heterogeneous information network embedding for recommendation[J].IEEE Transactions on Knowledge and Data Engineering，2018，31（2）：357-370.
[14] YIN J Y，GUO Y C，CHEN Y S.Heterogenous information network embedding based cross-domain recommendation system[C]//2019 International Conference on Data Mining Workshops（ICDMW），2019：362-369.
[15] CHEN L，LIU Y，ZHENG Z B，et al.Heterogeneous neural attentive factorization machine for rating prediction[C]//Proceedings of the 27th ACM International Conference on Information and Knowledge Management，2018：833-842.
[16] HU B B，SHI C，ZHAO W X，et al.Leveraging meta-path based context for top-n recommendation with a neural co-attention model[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining，2018：1531-1540.
[17] SUN Y Z，HAN J W.Mining heterogeneous information networks：principles and methodologies[J].Synthesis Lectures on Data Mining and Knowledge Discovery，2012，3（2）：1-159.
[18] ZHANG D K，YIN J，ZHU X Q，et al.Network representation learning：a survey[J].IEEE Transactions on Big Data，2018，6（1）：3-28.
[19] CUI P，WANG X，PEI J，et al.A survey on network embedding[J].IEEE Transactions on Knowledge and Data Engineering，2018，31（5）：833-852.
[20] SUN Y Z，NORICK B，HAN J W，et al.Pathselclus：integrating meta-path selection with user-guided object clustering in heterogeneous information networks[J].ACM Transactions on Knowledge Discovery from Data，2013，7（3）：1-23.
[21] GROVEK A，LESKOVE J.node2vec：scalable feature learning for networks[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining，2016：855-864.
[22] WANG X，JI H，SHI C，et al.Heterogeneous graph attention network[C]//The World Wide Web Conference，2019：2022-2032.
[23] MNIH A，SALAKHUTDINOV R R.Probabilistic matrix factorization[C]//Advances in Neural Information Processing Systems，2008：1257-1264.
[24] MA H，ZHOU D Y，LIU C，et al.Recommender systems with social regularization[C]//Proceedings of the Fourth ACM International Conference on Web Search and Data Mining，2011：287-296.
[25] ZHAO Z Y，ZHANG X J，ZHOU H，et al.HetNERec：heterogeneous network embedding based recommendation[J].Knowledge-Based Systems，2020，204（8）：106218.