Directional and Explainable Serendipity Recommendation

Serendipity recommendation has attracted more and more attention in recent years; it is committed to providing recommendations which could not only cater to users’ demands but also broaden their horizons. However, existing approaches usually measure user-item relevance with a scalar instead of a vector, ignoring user preference direction, which increases the risk of unrelated recommendations. In addition, reasonable explanations increase users’ trust and acceptance, but there is no work to provide explanations for serendipitous recommendations. To address these limitations, we propose a Directional and Explainable Serendipity Recommendation method named DESR. Specifically, we extract users’ long-term preferences with an unsupervised method based on GMM (Gaussian Mixture Model) and capture their short-term demands with the capsule network at first. Then, we propose the serendipity vector to combine long-term preferences with short-term demands and generate directionally serendipitous recommendations with it. Finally, a back-routing scheme is exploited to offer explanations. Extensive experiments on real-world datasets show that DESR could effectively improve the serendipity and explainability, and give impetus to the diversity, compared with existing serendipity-based methods.

[1]  Yanfang Ye,et al.  Heterogeneous Graph Attention Network , 2019, WWW.

[2]  Shuaiqiang Wang,et al.  Challenges of Serendipity in Recommender Systems , 2016, WEBIST.

[3]  Jie Wu,et al.  Trust Evaluation in Online Social Networks Using Generalized Network Flow , 2016, IEEE Transactions on Computers.

[4]  Geoffrey E. Hinton,et al.  Matrix capsules with EM routing , 2018, ICLR.

[5]  Wenjun Jiang,et al.  Brand purchase prediction based on time‐evolving user behaviors in e‐commerce , 2018, Concurr. Comput. Pract. Exp..

[6]  Mouzhi Ge,et al.  Beyond accuracy: evaluating recommender systems by coverage and serendipity , 2010, RecSys '10.

[7]  Filip Radlinski,et al.  Transparent, Scrutable and Explainable User Models for Personalized Recommendation , 2019, SIGIR.

[8]  Shuaiqiang Wang,et al.  A survey of serendipity in recommender systems , 2016, Knowl. Based Syst..

[9]  Kazjon Grace,et al.  Surprise Me If You Can: Serendipity in Health Information , 2018, CHI.

[10]  Geoffrey E. Hinton,et al.  Dynamic Routing Between Capsules , 2017, NIPS.

[11]  Jie Wu,et al.  HAES: A New Hybrid Approach for Movie Recommendation with Elastic Serendipity , 2019, CIKM.

[12]  Geoffrey E. Hinton,et al.  Transforming Auto-Encoders , 2011, ICANN.

[13]  Krzysztof Goczyla,et al.  Serendipitous Recommendations Through Ontology-Based Contextual Pre-filtering , 2017, BDAS.

[14]  Jie Wu,et al.  On Selecting Recommenders for Trust Evaluation in Online Social Networks , 2015, ACM Trans. Internet Techn..

[15]  Sahin Albayrak,et al.  User-centric evaluation of a K-furthest neighbor collaborative filtering recommender algorithm , 2013, CSCW.

[16]  Xiaoyu Zhang,et al.  Dressing as a Whole: Outfit Compatibility Learning Based on Node-wise Graph Neural Networks , 2019, WWW.

[17]  Li Chen,et al.  How Serendipity Improves User Satisfaction with Recommendations? A Large-Scale User Evaluation , 2019, WWW.

[18]  Jie Wu,et al.  Understanding Graph-Based Trust Evaluation in Online Social Networks , 2016, ACM Comput. Surv..

[19]  Xi Niu.  An Adaptive Recommender System for Computational Serendipity , 2018, ICTIR.

[20]  Julian J. McAuley,et al.  Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering , 2016, WWW.

[21]  Yuan He,et al.  Graph Neural Networks for Social Recommendation , 2019, WWW.

[22]  Lora Aroyo,et al.  SIRUP: Serendipity In Recommendations via User Perceptions , 2017, IUI.

[23]  Gaurav Pandey,et al.  Recommending Serendipitous Items using Transfer Learning , 2018, CIKM.

[24]  Xing Xie,et al.  Attention-driven Factor Model for Explainable Personalized Recommendation , 2018, SIGIR.

[25]  Yoshua Bengio,et al.  Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling , 2014, ArXiv.

[26]  Min Yang,et al.  NAIRS: A Neural Attentive Interpretable Recommendation System , 2019, WSDM.

[27]  Jürgen Schmidhuber,et al.  Learning to Forget: Continual Prediction with LSTM , 2000, Neural Computation.

[28]  Jure Leskovec,et al.  Hierarchical Temporal Convolutional Networks for Dynamic Recommender Systems , 2019, WWW.

[29]  Carl E. Rasmussen,et al.  The Infinite Gaussian Mixture Model , 1999, NIPS.

[30]  Anton van den Hengel,et al.  Image-Based Recommendations on Styles and Substitutes , 2015, SIGIR.

[31]  R. Redner,et al.  Mixture densities, maximum likelihood, and the EM algorithm , 1984 .

[32]  Ed H. Chi,et al.  Towards Neural Mixture Recommender for Long Range Dependent User Sequences , 2019, WWW.

[33]  Yan Wang,et al.  ProNE: Fast and Scalable Network Representation Learning , 2019, IJCAI.

[34]  James McInerney,et al.  Explore, exploit, and explain: personalizing explainable recommendations with bandits , 2018, RecSys.

[35]  F. Maxwell Harper,et al.  The MovieLens Datasets: History and Context , 2016, TIIS.

[36]  Daniele Quercia,et al.  Auralist: introducing serendipity into music recommendation , 2012, WSDM '12.

[37]  Lukás Burget,et al.  Recurrent neural network based language model , 2010, INTERSPEECH.

[38]  Yiqun Liu,et al.  Jointly Learning Explainable Rules for Recommendation with Knowledge Graph , 2019, WWW.

[39]  Haibin Cheng,et al.  Real-time Personalization using Embeddings for Search Ranking at Airbnb , 2018, KDD.

[40]  Yongfeng Zhang,et al.  Dynamic Explainable Recommendation Based on Neural Attentive Models , 2019, AAAI.

[41]  Jie Wu,et al.  Forming Opinions via Trusted Friends: Time-Evolving Rating Prediction Using Fluid Dynamics , 2016, IEEE Transactions on Computers.

[42]  Shuaiqiang Wang,et al.  How does serendipity affect diversity in recommender systems? A serendipity-oriented greedy algorithm , 2018, Computing.