ATBRG: Adaptive Target-Behavior Relational Graph Network for Effective Recommendation

Feng Yufei
Feng Yufei
Hu Binbin
Hu Binbin
Liu Qingwen
Liu Qingwen

SIGIR '20: The 43rd International ACM SIGIR conference on research and development in Information Retrieval Virtual Event China July, 2020, pp. 2231-2240, 2020.

Cited by: 0|Bibtex|Views64|DOI:https://doi.org/10.1145/3397271.3401428
EI
Other Links: arxiv.org|dl.acm.org|dblp.uni-trier.de|academic.microsoft.com
Weibo:
We propose a new framework named Adaptive Target-Behavior Relational Graph network to effectively capture structural relations of target user-item pairs over knowledge graph

Abstract:

Recommender system (RS) devotes to predicting user preference to a given item and has been widely deployed in most web-scale applications. Recently, knowledge graph (KG) attracts much attention in RS due to its abundant connective information. Existing methods either explore independent meta-paths for user-item pairs over KG, or employ gr...More

Code:

Data:

0
Introduction
  • In the era of information overload, recommender system (RS), which aims to match diverse user interests with tremendous resource items, are widely deployed in various online services, including e-commerce [17, 27, 32], social media [3, 33] and news [4, 21].
  • Traditional recommendation methods, e.g., matrix factorization [11], mainly learn an effective preference prediction function using historical user-item interaction records.
  • Despite effectiveness, these methods suffer from cold-start problem due to data sparsity.
  • With the rapid development of web services, some approaches [8, 9] are proposed to incorporate various auxiliary data for improving recommendation performance.
  • Due to its abundant information, current recommender systems mainly aim to incorporate KG to enrich representations of users and items and promote the interpretability of recommendations
Highlights
  • In the era of information overload, recommender system (RS), which aims to match diverse user interests with tremendous resource items, are widely deployed in various online services, including e-commerce [17, 27, 32], social media [3, 33] and news [4, 21]
  • We aim to distill the original overinformative knowledge graph (KG) into recommendation in a more effective way, which is expected to satisfy the following key properties: (1) Targetbehavior: we hang on the novel insight that an effective KG base recommendation should produce semantic sub-graph to adapt for each target user-item pair, with the aim of capturing the underlying mutual effect characterized by KG (L1); (2) Adaptive: distinct from random sampling on the whole KG, our idea is to follow the adaptive principle for the sub-graph construction, which adaptively preserves useful information connecting user behaviors and target item over the KG, driving our model to provide more effective recommendation (L2); (3) Relational: the model architecture should be designed to relation-aware in order to consider the rich relations among user behaviors and target item over KG (L3)
  • The framework is shown in Fig. 2, which is composed of two modules: (1) To effectively extract structural relational knowledge for recommendation, we propose to construct the adaptive target-behavior relational graph for the given target user-item pair over knowledge graph, where the graph connect and graph prune techniques help mine high-order connective structure in an automatic manner; (2) To jointly distill such a relational graph and rich relations among user behaviors in an end-to-end framework, we elaborate on the model design of Adaptive Target-Behavior Relational Graph Network (ATBRG), which propagates user preference on the sub-graph with relation-aware extractor layer and representation activation layer
  • To verify the effectiveness of our proposed framework ATBRG in the real-world settings, ATBRG has been deployed in the popular recommendation scenario of Taobao APP
  • To effectively characterize the structure relations over KG, we propose the graph connect and graph prune techniques to construct adaptive target-behavior relational graph
  • We argue that above two strategies only achieve the suboptimal performance for recommendation
  • We will consider applying causal inference in KG to improve the interpretability of recommender system
Results
  • To verify the effectiveness of the proposed framework ATBRG in the real-world settings, ATBRG has been deployed in the popular recommendation scenario of Taobao APP.
  • Users give implicit feedback to the recommended items provided by the recommender system; 2) Offline training.
  • When the user accesses Taobao APP, some candidates items are generated by the pipelines before real-time prediction (RTP) service.
  • The promotion of recommendation performance verifies the effectiveness of the proposed framework ATBRG
Conclusion
  • The authors propose a novel framework ATBRG for knowledge aware recommendation.
  • The authors elaborate on the model design of ATBRG, equipped with relation-aware extractor layer and representation activation layer, which aims to take full advantage of structural connective knowledge for recommendation.
  • Extensive experiments on both industrial and benchmark datasets demonstrate the effectiveness of the framework compared to several state-of-the-art methods.
  • The authors will consider applying causal inference in KG to improve the interpretability of recommender system
Summary
  • Introduction:

    In the era of information overload, recommender system (RS), which aims to match diverse user interests with tremendous resource items, are widely deployed in various online services, including e-commerce [17, 27, 32], social media [3, 33] and news [4, 21].
  • Traditional recommendation methods, e.g., matrix factorization [11], mainly learn an effective preference prediction function using historical user-item interaction records.
  • Despite effectiveness, these methods suffer from cold-start problem due to data sparsity.
  • With the rapid development of web services, some approaches [8, 9] are proposed to incorporate various auxiliary data for improving recommendation performance.
  • Due to its abundant information, current recommender systems mainly aim to incorporate KG to enrich representations of users and items and promote the interpretability of recommendations
  • Objectives:

    The authors aim to distill the original overinformative KG into recommendation in a more effective way, which is expected to satisfy the following key properties: (1) Targetbehavior: the authors hang on the novel insight that an effective KG base recommendation should produce semantic sub-graph to adapt for each target user-item pair, with the aim of capturing the underlying mutual effect characterized by KG (L1); (2) Adaptive: distinct from random sampling on the whole KG, the idea is to follow the adaptive principle for the sub-graph construction, which adaptively preserves useful information connecting user behaviors and target item over the KG, driving the model to provide more effective recommendation (L2); (3) Relational: the model architecture should be designed to relation-aware in order to consider the rich relations among user behaviors and target item over KG (L3).
  • Given a knowledge graph G with historical interaction records H , for each user-item pair ⟨u, i⟩, the authors aim to predict probability yui that user u would click item i
  • Results:

    To verify the effectiveness of the proposed framework ATBRG in the real-world settings, ATBRG has been deployed in the popular recommendation scenario of Taobao APP.
  • Users give implicit feedback to the recommended items provided by the recommender system; 2) Offline training.
  • When the user accesses Taobao APP, some candidates items are generated by the pipelines before real-time prediction (RTP) service.
  • The promotion of recommendation performance verifies the effectiveness of the proposed framework ATBRG
  • Conclusion:

    The authors propose a novel framework ATBRG for knowledge aware recommendation.
  • The authors elaborate on the model design of ATBRG, equipped with relation-aware extractor layer and representation activation layer, which aims to take full advantage of structural connective knowledge for recommendation.
  • Extensive experiments on both industrial and benchmark datasets demonstrate the effectiveness of the framework compared to several state-of-the-art methods.
  • The authors will consider applying causal inference in KG to improve the interpretability of recommender system
Tables
  • Table1: Notations
  • Table2: Statistics of datasets
  • Table3: Overall performance comparison w.r.t. AUC (bold: best; underline: runner-up)
  • Table4: Effect of the representation activation layer and relation-aware mechanism
  • Table5: Effect of the depth of neighbor
  • Table6: Effect of different aggregators
Download tables as Excel
Related work
  • In this section, we review the most related studies in behavior based and knowledge aware recommendation.

    2.1 Behavior based recommendation

    In the early stage of recommendation, researchers focus on recommending a suitable list of items based on historical user-item interaction records. In particular, a series of matrix factorization based methods [11] have been proposed to infer user preference towards items through learning latent representations of users and items. Due to the ability of modeling complex interaction between users and items, deep neural network based methods (e.g., YoutubeNet [3], DeepFM [7]) are widely adopted in industrial recommender systems, and reveal the remarkable strength of incorporating various context information (e.g., user profile and item attributes).

    In the online e-commerce systems, we are particularly interested in user’s historical behaviors, which implies rich information for inferring user preference. Hence, how to effectively characterize the relationships between user behaviors and target item remains a continuous research topic. DIN [32] adaptively learns the representation of user interests from historical behaviors w.r.t. the target item by the attention mechanism. Inspired by DIN, the majority of following up works inherit this kind of paradigm. GIN [12] mines user intention based on co-occurrence commodity graph in the end-to-end fashion. ATRANK [30] proposes an attentionbased behavior modeling framework to model users’ heterogeneous behaviors. DIEN [31] and SDM [14] devote to capturing users’ temporal interests and modeling their sequential relations. DSIN [5] focuses on capturing the relationships of users’ inter-session and intra-session behaviors. MIMN [15] and HPMN [16] apply the neural turing machine to model users’ lifelong sequential behaviors. Besides these improvements, knowledge graph, consisting of various semantics and relations, emerges as an assistant to describe relationships between user behaviors and target item.
Reference
  • Dzmitry Bahdanau, Kyunghyun Cho, and Yoshua Bengio. 2015. Neural machine translation by jointly learning to align and translate. In ICLR.
    Google ScholarFindings
  • Yixin Cao, Xiang Wang, Xiangnan He, Zikun Hu, and Tat-Seng Chua. 2019. Unifying knowledge graph learning and recommendation: Towards a better understanding of user preferences. In WWW. 151–161.
    Google ScholarLocate open access versionFindings
  • Covington, Paul, Adams, Jay, Sargin, and Emre. 2016. Deep neural networks for youtube recommendations. In RecSys. 191–198.
    Google ScholarFindings
  • Abhinandan S Das, Mayur Datar, Ashutosh Garg, and Shyam Rajaram. 2007. Google news personalization: scalable online collaborative filtering. In WWW. 271–280.
    Google ScholarLocate open access versionFindings
  • Yufei Feng, Fuyu Lv, Weichen Shen, Menghan Wang, Fei Sun, Yu Zhu, and Keping Yang. 2019. Deep Session Interest Network for Click-Through Rate Prediction. In IJCAI. 2301–2307.
    Google ScholarFindings
  • Liyu Gong and Qiang Cheng. 2019. Exploiting edge features for graph neural networks. In CVPR. 9211–9219.
    Google ScholarFindings
  • Huifeng Guo, Ruiming Tang, Yunming Ye, Zhenguo Li, and Xiuqiang He. 201DeepFM: a factorization-machine based neural network for CTR prediction. In IJCAI. 1725–1731.
    Google ScholarFindings
  • Binbin Hu, Chuan Shi, Wayne Xin Zhao, and Philip S Yu. 201Leveraging meta-path based context for top-n recommendation with a neural co-attention model. In SIGKDD. 1531–1540.
    Google ScholarFindings
  • Jin Huang, Wayne Xin Zhao, Hongjian Dou, Ji-Rong Wen, and Edward Y Chang. 2018. Improving sequential recommendation with knowledge-enhanced memory networks. In SIGIR. 505–514.
    Google ScholarLocate open access versionFindings
  • Thomas N Kipf and Max Welling. 2017. Semi-supervised classification with graph convolutional networks. In ICLR.
    Google ScholarFindings
  • Yehuda Koren, Robert Bell, and Chris Volinsky. 2009. Matrix factorization techniques for recommender systems. Computer 42, 8 (2009), 30–37.
    Google ScholarLocate open access versionFindings
  • Feng Li, Zhenrui Chen, Pengjie Wang, Yi Ren, Di Zhang, and Xiaoyu Zhu. 2019. Graph Intention Network for Click-through Rate Prediction in Sponsored Search. In SIGIR. 961–964.
    Google ScholarLocate open access versionFindings
  • Xusheng Luo, Luxin Liu, Yonghua Yang, Le Bo, Yuanpeng Cao, Jinhang Wu, Qiang Li, Keping Yang, and Kenny Q. Zhu. 2020. AliCoCo: Alibaba E-commerce Cognitive Concept Net. In SIGMOD.
    Google ScholarLocate open access versionFindings
  • Fuyu Lv, Taiwei Jin, Changlong Yu, Fei Sun, Quan Lin, Keping Yang, and Wilfred Ng. 2019. SDM: Sequential deep matching model for online large-scale recommender system. In Proceedings of the 28th ACM International Conference on Information and Knowledge Management. 2635–2643.
    Google ScholarLocate open access versionFindings
  • Qi Pi, Weijie Bian, Guorui Zhou, Xiaoqiang Zhu, and Kun Gai. 2019. Practice on
    Google ScholarFindings
  • Kan Ren, Jiarui Qin, Yuchen Fang, Weinan Zhang, Lei Zheng, Weijie Bian, Guorui Zhou, Jian Xu, Yong Yu, Xiaoqiang Zhu, and Kun Gai. 2019. Lifelong Sequential
    Google ScholarFindings
  • Badrul Munir Sarwar, George Karypis, Joseph A Konstan, John Riedl, et al. 2001. Item-based collaborative filtering recommendation algorithms. In WWW. 285– 295.
    Google ScholarLocate open access versionFindings
  • Wei Shen, Jianyong Wang, and Jiawei Han. 2014. Entity linking with a knowledge base: Issues, techniques, and solutions. IEEE Transactions on Knowledge and Data Engineering 27, 2 (2014), 443–460.
    Google ScholarLocate open access versionFindings
  • Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Lio, and Yoshua Bengio. 2018. Graph attention networks. In ICLR.
    Google ScholarFindings
  • Hongwei Wang, Fuzheng Zhang, Jialin Wang, Miao Zhao, Wenjie Li, Xing Xie, and Minyi Guo. 2018. Ripplenet: Propagating user preferences on the knowledge graph for recommender systems. In CIKM. 417–426.
    Google ScholarLocate open access versionFindings
  • Hongwei Wang, Fuzheng Zhang, Xing Xie, and Minyi Guo. 2018. DKN: Deep knowledge-aware network for news recommendation. In WWW. 1835–1844.
    Google ScholarFindings
  • Hongwei Wang, Miao Zhao, Xing Xie, Wenjie Li, and Minyi Guo. 2019. Knowledge graph convolutional networks for recommender systems. In WWW. 3307–3313.
    Google ScholarFindings
  • Xiang Wang, Xiangnan He, Yixin Cao, Meng Liu, and Tat-Seng Chua. 2019. KGAT: Knowledge Graph Attention Network for Recommendation. In SIGKDD. 950–958.
    Google ScholarLocate open access versionFindings
  • Xiang Wang, Dingxian Wang, Canran Xu, Xiangnan He, Yixin Cao, and Tat-Seng Chua. 2019. Explainable reasoning over knowledge graphs for recommendation. In AAAI. 5329–5336.
    Google ScholarFindings
  • Xiang Wang, Dingxian Wang, Canran Xu, Xiangnan He, Yixin Cao, and Tat-Seng
    Google ScholarFindings
  • Yikun Xian, Zuohui Fu, S Muthukrishnan, Gerard De Melo, and Yongfeng Zhang.
    Google ScholarFindings
  • 2019. Reinforcement knowledge graph reasoning for explainable recommendation. In SIGIR. 285–294.
    Google ScholarLocate open access versionFindings
  • [27] Fengli Xu, Jianxun Lian, Zhenyu Han, Yong Li, Yujian Xu, and Xing Xie. 2019.
    Google ScholarFindings
  • [28] Fuzheng Zhang, Nicholas Jing Yuan, Defu Lian, Xing Xie, and Wei-Ying Ma. 2016. Collaborative knowledge base embedding for recommender systems. In SIGKDD. 353–362.
    Google ScholarLocate open access versionFindings
  • [29] Wayne Xin Zhao, Gaole He, Kunlin Yang, Hongjian Dou, Jin Huang, Siqi Ouyang, and Ji-Rong Wen. 2019. KB4Rec: A Data Set for Linking Knowledge Bases with Recommender Systems. Data Intelligence 1, 2 (2019), 121–136.
    Google ScholarLocate open access versionFindings
  • [30] Chang Zhou, Jinze Bai, Junshuai Song, Xiaofei Liu, Zhengchao Zhao, Xiusi Chen, and Jun Gao. 2018. ATRank: An attention-based user behavior modeling framework for recommendation. In AAAI. 4564–4571.
    Google ScholarFindings
  • [31] Guorui Zhou, Na Mou, Ying Fan, Qi Pi, Weijie Bian, Chang Zhou, Xiaoqiang Zhu, and Kun Gai. 2019. Deep interest evolution network for click-through rate prediction. In AAAI. 5941–5948.
    Google ScholarFindings
  • [32] Guorui Zhou, Xiaoqiang Zhu, Chenru Song, Ying Fan, Han Zhu, Xiao Ma, Yanghui Yan, Junqi Jin, Han Li, and Kun Gai. 2018. Deep interest network for click-through rate prediction. In SIGKDD. 1059–1068.
    Google ScholarFindings
  • [33] Xiangmin Zhou, Dong Qin, Lei Chen, and Yanchun Zhang. 2019. Real-time context-aware social media recommendation. The VLDB Journal 28, 2 (2019), 197–219.
    Google ScholarLocate open access versionFindings
Full Text
Your rating :
0

 

Tags
Comments