CRFR: Improving Conversational Recommender Systems via Flexible Fragments Reasoning on Knowledge Graphs

Although paths of user interests shift in knowledge graphs (KGs) can benefit conversational recommender systems (CRS), explicit reasoning on KGs has not been well considered in CRS, due to the complex of high-order and incomplete paths. We propose CRFR, which effectively does explicit multi-hop reasoning on KGs with a conversational context-based reinforcement learning model. Considering the incompleteness of KGs, instead of learning single complete reasoning path, CRFR flexibly learns multiple reasoning fragments which are likely contained in the complete paths of interests shift. A fragments-aware unified model is then designed to fuse the fragments information from item-oriented and concept-oriented KGs to enhance the CRS response with entities and words from the fragments. Extensive experiments demonstrate CRFR’s SOTA performance on recommendation, conversation and conversation interpretability.

[1]  Ji-rong Wen,et al.  Adapting User Preference to Online Feedback in Multi-round Conversational Recommendation , 2021, WSDM.

[2]  M. de Rijke,et al.  Advances and Challenges in Conversational Recommender Systems: A Survey , 2021, AI Open.

[3]  Mihael Arcan,et al.  Suggest me a movie for tonight: Leveraging Knowledge Graphs for Conversational Recommendation , 2020, COLING.

[4]  Minlie Huang,et al.  Bridging the Gap between Conversational Reasoning and Interactive Recommendation , 2020, ArXiv.

[5]  Wayne Xin Zhao,et al.  Towards Topic-Guided Conversational Recommender System , 2020, COLING.

[6]  Wei Zhang,et al.  Dynamic Anticipation and Completion for Multi-Hop Reasoning over Sparse Knowledge Graph , 2020, EMNLP.

[7]  Weiyan Shi,et al.  INSPIRED: Toward Sociable Recommendation Dialog Systems , 2020, EMNLP.

[8]  Scott Sanner,et al.  A Ranking Optimization Approach to Latent Linear Critiquing for Conversational Recommender Systems , 2020, RecSys.

[9]  Kun Zhou,et al.  Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion , 2020, KDD.

[10]  Xiangnan He,et al.  Interactive Path Reasoning on Graph for Conversational Recommendation , 2020, KDD.

[11]  Xing Xie,et al.  Towards Explainable Conversational Recommendation , 2020, IJCAI.

[12]  Zheng-Yu Niu,et al.  Conversational Graph Grounded Policy Learning for Open-Domain Conversation Generation , 2020, ACL.

[13]  Evangelos Kanoulas,et al.  Towards Question-based Recommender Systems , 2020, SIGIR.

[14]  Zheng-Yu Niu,et al.  Towards Conversational Recommendation over Multi-Type Dialogs , 2020, ACL.

[15]  Scott Sanner,et al.  Latent Linear Critiquing for Conversational Recommender Systems , 2020, WWW.

[16]  D. Jannach,et al.  A Survey on Conversational Recommender Systems , 2020, ACM Comput. Surv..

[17]  Xiangnan He,et al.  Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems , 2020, WSDM.

[18]  Pararth Shah,et al.  Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue , 2019, EMNLP.

[19]  Hongxia Yang,et al.  Towards Knowledge-Based Recommender Dialog System , 2019, EMNLP.

[20]  Seungwhan Moon,et al.  OpenDialKG: Explainable Conversational Reasoning with Attention-based Walks over Knowledge Graphs , 2019, ACL.

[21]  Tat-Seng Chua,et al.  Deep Conversational Recommender in Travel , 2019, ArXiv.

[22]  Christopher Joseph Pal,et al.  Towards Deep Conversational Recommendations , 2018, NeurIPS.

[23]  Xu Chen,et al.  Towards Conversational Search and Recommendation: System Ask, User Respond , 2018, CIKM.

[24]  Yi Zhang,et al.  Conversational Recommender System , 2018, SIGIR.

[25]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[26]  Max Welling,et al.  Modeling Relational Data with Graph Convolutional Networks , 2017, ESWC.

[27]  Catherine Havasi,et al.  ConceptNet 5.5: An Open Multilingual Graph of General Knowledge , 2016, AAAI.

[28]  Joelle Pineau,et al.  Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models , 2015, AAAI.

[29]  Yoshua Bengio,et al.  A Hierarchical Recurrent Encoder-Decoder for Generative Context-Aware Query Suggestion , 2015, CIKM.

[30]  Yoon Kim,et al.  Convolutional Neural Networks for Sentence Classification , 2014, EMNLP.

[31]  Jens Lehmann,et al.  DBpedia - A crystallization point for the Web of Data , 2009, J. Web Semant..

[32]  Xianghua Xie,et al.  Graph Convolutional Neural Network , 2016, British Machine Vision Conference.

[33]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.