Reinforcement knowledge graph reasoning for explainable recommendation

Yikun Xian, Zuohui Fu, S. Muthukrishnan, Gerard De Melo, Yongfeng Zhang

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Recent advances in personalized recommendation have sparked great interest in the exploitation of rich structured information provided by knowledge graphs. Unlike most existing approaches that only focus on leveraging knowledge graphs for more accurate recommendation, we perform explicit reasoning with knowledge for decision making so that the recommendations are generated and supported by an interpretable causal inference procedure. To this end, we propose a method called Policy-Guided Path Reasoning (PGPR), which couples recommendation and interpretability by providing actual paths in a knowledge graph. Our contributions include four aspects. We first highlight the significance of incorporating knowledge graphs into recommendation to formally define and interpret the reasoning process. Second, we propose a reinforcement learning (RL) approach featuring an innovative soft reward strategy, user-conditional action pruning and a multi-hop scoring function. Third, we design a policy-guided graph search algorithm to efficiently and effectively sample reasoning paths for recommendation. Finally, we extensively evaluate our method on several large-scale real-world benchmark datasets, obtaining favorable results compared with state-of-the-art methods.

    Original languageEnglish (US)
    Title of host publicationSIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval
    PublisherAssociation for Computing Machinery, Inc
    Pages285-294
    Number of pages10
    ISBN (Electronic)9781450361729
    DOIs
    StatePublished - Jul 18 2019
    Event42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2019 - Paris, France
    Duration: Jul 21 2019Jul 25 2019

    Publication series

    NameSIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval

    Conference

    Conference42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2019
    CountryFrance
    CityParis
    Period7/21/197/25/19

    Fingerprint

    Reinforcement learning
    Reinforcement
    Recommendations
    Reasoning
    Decision making
    Graph in graph theory
    Path
    Graph Search
    Personalized Recommendation
    Causal Inference
    Graph Algorithms
    Interpretability
    Multi-hop
    Reinforcement Learning
    Pruning
    Scoring
    Reward
    Exploitation
    Search Algorithm
    Decision Making

    Keywords

    • Explainability
    • Knowledge Graphs
    • Recommendation System
    • Reinforcement Learning

    ASJC Scopus subject areas

    • Information Systems
    • Applied Mathematics
    • Software

    Cite this

    Xian, Y., Fu, Z., Muthukrishnan, S., De Melo, G., & Zhang, Y. (2019). Reinforcement knowledge graph reasoning for explainable recommendation. In SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 285-294). (SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval). Association for Computing Machinery, Inc. https://doi.org/10.1145/3331184.3331203

    Reinforcement knowledge graph reasoning for explainable recommendation. / Xian, Yikun; Fu, Zuohui; Muthukrishnan, S.; De Melo, Gerard; Zhang, Yongfeng.

    SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, 2019. p. 285-294 (SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval).

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Xian, Y, Fu, Z, Muthukrishnan, S, De Melo, G & Zhang, Y 2019, Reinforcement knowledge graph reasoning for explainable recommendation. in SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, Association for Computing Machinery, Inc, pp. 285-294, 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2019, Paris, France, 7/21/19. https://doi.org/10.1145/3331184.3331203
    Xian Y, Fu Z, Muthukrishnan S, De Melo G, Zhang Y. Reinforcement knowledge graph reasoning for explainable recommendation. In SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc. 2019. p. 285-294. (SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval). https://doi.org/10.1145/3331184.3331203
    Xian, Yikun ; Fu, Zuohui ; Muthukrishnan, S. ; De Melo, Gerard ; Zhang, Yongfeng. / Reinforcement knowledge graph reasoning for explainable recommendation. SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, 2019. pp. 285-294 (SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval).
    @inproceedings{de0b85c1b803483f89eefaa9444f2bb9,
    title = "Reinforcement knowledge graph reasoning for explainable recommendation",
    abstract = "Recent advances in personalized recommendation have sparked great interest in the exploitation of rich structured information provided by knowledge graphs. Unlike most existing approaches that only focus on leveraging knowledge graphs for more accurate recommendation, we perform explicit reasoning with knowledge for decision making so that the recommendations are generated and supported by an interpretable causal inference procedure. To this end, we propose a method called Policy-Guided Path Reasoning (PGPR), which couples recommendation and interpretability by providing actual paths in a knowledge graph. Our contributions include four aspects. We first highlight the significance of incorporating knowledge graphs into recommendation to formally define and interpret the reasoning process. Second, we propose a reinforcement learning (RL) approach featuring an innovative soft reward strategy, user-conditional action pruning and a multi-hop scoring function. Third, we design a policy-guided graph search algorithm to efficiently and effectively sample reasoning paths for recommendation. Finally, we extensively evaluate our method on several large-scale real-world benchmark datasets, obtaining favorable results compared with state-of-the-art methods.",
    keywords = "Explainability, Knowledge Graphs, Recommendation System, Reinforcement Learning",
    author = "Yikun Xian and Zuohui Fu and S. Muthukrishnan and {De Melo}, Gerard and Yongfeng Zhang",
    year = "2019",
    month = "7",
    day = "18",
    doi = "10.1145/3331184.3331203",
    language = "English (US)",
    series = "SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval",
    publisher = "Association for Computing Machinery, Inc",
    pages = "285--294",
    booktitle = "SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval",

    }

    TY - GEN

    T1 - Reinforcement knowledge graph reasoning for explainable recommendation

    AU - Xian, Yikun

    AU - Fu, Zuohui

    AU - Muthukrishnan, S.

    AU - De Melo, Gerard

    AU - Zhang, Yongfeng

    PY - 2019/7/18

    Y1 - 2019/7/18

    N2 - Recent advances in personalized recommendation have sparked great interest in the exploitation of rich structured information provided by knowledge graphs. Unlike most existing approaches that only focus on leveraging knowledge graphs for more accurate recommendation, we perform explicit reasoning with knowledge for decision making so that the recommendations are generated and supported by an interpretable causal inference procedure. To this end, we propose a method called Policy-Guided Path Reasoning (PGPR), which couples recommendation and interpretability by providing actual paths in a knowledge graph. Our contributions include four aspects. We first highlight the significance of incorporating knowledge graphs into recommendation to formally define and interpret the reasoning process. Second, we propose a reinforcement learning (RL) approach featuring an innovative soft reward strategy, user-conditional action pruning and a multi-hop scoring function. Third, we design a policy-guided graph search algorithm to efficiently and effectively sample reasoning paths for recommendation. Finally, we extensively evaluate our method on several large-scale real-world benchmark datasets, obtaining favorable results compared with state-of-the-art methods.

    AB - Recent advances in personalized recommendation have sparked great interest in the exploitation of rich structured information provided by knowledge graphs. Unlike most existing approaches that only focus on leveraging knowledge graphs for more accurate recommendation, we perform explicit reasoning with knowledge for decision making so that the recommendations are generated and supported by an interpretable causal inference procedure. To this end, we propose a method called Policy-Guided Path Reasoning (PGPR), which couples recommendation and interpretability by providing actual paths in a knowledge graph. Our contributions include four aspects. We first highlight the significance of incorporating knowledge graphs into recommendation to formally define and interpret the reasoning process. Second, we propose a reinforcement learning (RL) approach featuring an innovative soft reward strategy, user-conditional action pruning and a multi-hop scoring function. Third, we design a policy-guided graph search algorithm to efficiently and effectively sample reasoning paths for recommendation. Finally, we extensively evaluate our method on several large-scale real-world benchmark datasets, obtaining favorable results compared with state-of-the-art methods.

    KW - Explainability

    KW - Knowledge Graphs

    KW - Recommendation System

    KW - Reinforcement Learning

    UR - http://www.scopus.com/inward/record.url?scp=85073797147&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85073797147&partnerID=8YFLogxK

    U2 - 10.1145/3331184.3331203

    DO - 10.1145/3331184.3331203

    M3 - Conference contribution

    AN - SCOPUS:85073797147

    T3 - SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval

    SP - 285

    EP - 294

    BT - SIGIR 2019 - Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval

    PB - Association for Computing Machinery, Inc

    ER -