Abstract：Current music recommendation models can mine the general preference between users and music,which are unable to distinguish the differential preference of different users towards the same song.Therefore,we propose a hierarchical attention representation model (HARM) to improve the music recommendation quality.HARM utilizes attributes of users and contents of music to learn music representation from the perspective of multi-dimension and mine the preference relationship between user and music.In order to mine users' differential preference on music's multi-field feature,a user feature-dependent attention network is designed.In addition,in order to mine the impacts of different historical behavior on user preference and learn sequential dependency of user's,a music-dependent attention network is designed.Finally,recommendation is generated by using a softmax function to calculate the preference distribution of users on candidate songs.The experimental results on 30Music and MIGU datasets shows that,comparing with the existent recommendation models,HARM can gain significant improvement on Recall and MRR.
[1] Turrin R,Quadrana M,Condorelli A,et al.30music listening and playlists dataset[A].Proceedings of the 9th ACM Conference on Recommender Systems[C].New York:Association for Computing Machinery,2015.1-4.
[2] 任开旭,王玉龙,刘同存,等.融合多维语义表示的概率矩阵分解模型[J].电子学报,2019,47(9):1848-1854. Ren K X,Wang Y L,Liu T C,et al.A probabilistic matrix factorization model based on multidimensional semantic representation learning[J].Acta Electronica Sinica,2019,47(9):1848-1854.(in Chinese)
[3] Rendle S,Freudenthaler C,Gantner Z,et al.BPR:Bayesian personalized ranking from implicit feedback[A].Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence[C].Virginia:AUAI Press,2009.452-461.
[4] Van den Oord A,Dieleman S,Schrauwen B.Deep content-based music recommendation[A].Advances in Neural Information Processing Systems[C].New York:NIPS,2013.2643-2651.
[5] Dias R,Fonseca M J.Improving music recommendation in session-based collaborative filtering by using temporal context[A].2013 IEEE 25th International Conference on Tools with Artificial Intelligence[C].Piscataway:IEEE,2013.783-788.
[6] Wang D,Deng S,Xu G.Sequence-based context-aware music recommendation[J].Information Retrieval Journal,2018,21(2-3):230-252.
[7] He R,Kang W C,McAuley J.Translation-based recommendation[A].Proceedings of the Eleventh ACM Conference on Recommender Systems[C].New York:Association for Computing Machinery,2017.161-169.
[8] He R,McAuley J.Fusing similarity models with markov chains for sparse sequential recommendation[A].2016 IEEE 16th International Conference on Data Mining (ICDM)[C].Washington:IEEE Computer Society,2016.191-200.
[9] Tavakol M,Brefeld U.Factored MDPs for detecting topics of user sessions[A].Proceedings of the 8th ACM Conference on Recommender Systems[C].New York:Association for Computing Machinery,2014.33-40.
[10] Twardowski B.Modelling contextual information in session-aware recommender systems with neural networks[A].Proceedings of the 10th ACM Conference on Recommender Systems[C].New York:Association for Computing Machinery,2016.273-276.
[11] Kabbur S,Ning X,Karypis G.Fism:factored item similarity models for top-n recommender systems[A].Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining[C].New York:Association for Computing Machinery,2013.659-667.
[12] 潘博,于重重,张青川,等.基于词性与词序的相关因子训练的word2vec改进模型[J].电子学报,2018,46(8):1976-1982. Pan B,Yu C C,Zhang Q C,et al.The improved model for word2vec based on part of speech and word order[J].Acta Electronica Sinica,2018,46(8):1976-1982.(in Chinese)
[13] Hidasi B,Karatzoglou A,Baltrunas L,et al.Session-based recommendations with Recurrent Neural Networks[OL].https://arxiv.org/abs/1511.06939,2015.
[14] Tan Y K,Xu X,Liu Y.Improved recurrent neural networks for session-based recommendations[A].Proceedings of the 1st Workshop on Deep Learning for Recommender Systems[C].New York:Association for Computing Machinery,2016.17-22.
[15] Quadrana M,Karatzoglou A,Personalizing session-based recommendations with hierarchical recurrent neural networks[A].Proceedings of the Eleventh ACM Conference on Recommender Systems[C].New York:Association for Computing Machinery,2017.130-137.
[16] Chatzis S P,Christodoulou P,Andreou A S.Recurrent latent variable networks for session-based recommendation[A].Proceedings of the 2nd Workshop on Deep Learning for Recommender Systems[C].New York:Association for Computing Machinery,2017.38-45.
[17] Wu C,Yan M.Session-aware information embedding for e-commerce product recommendation[A].Proceedings of the 2017 ACM on Conference on Information and Knowledge Management[C].New York:Association for Computing Machinery,2017.2379-2382.
[18] Song Y,Elkahky A M,He X.Multi-rate deep learning for temporal recommendation[A].Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval[C].New York:Association for Computing Machinery,2016.909-912.
[19] Tang J,Wang K.Personalized top-n sequential recommendation via convolutional sequence embedding[A].Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining[C].New York:Association for Computing Machinery,2018.565-573.
[20] 冯建周,宋沙沙,王元卓,等.基于改进注意力机制的实体关系抽取方法[J].电子学报,2019,47(8):1692-1700. Feng J Z,Song S S,Wang Y Z,et al.Entity relation extraction based on improved attention mechanism[J].Acta Electronica Sinica,2019,47(8):1692-1700.(in Chinese)
[21] Li J,Ren P,Chen Z,et al.Neural attentive session-based recommendation[A].Proceedings of the 2017 ACM on Conference on Information and Knowledge Management[C].New York:Association for Computing Machinery,2017.1419-1428.
[22] Liu Q,Zeng Y,Mokhosi R,et al.STAMP:short-term attention/memory priority model for session-based recommendation[A].Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining[C].New York:Association for Computing Machinery,2018.1831-1839.
[23] Sachdeva N,Gupta K,Pudi V.Attentive neural architecture incorporating song features for music recommendation[A].Proceedings of the 12th ACM Conference on Recommender Systems[C].New York:Association for Computing Machinery,2018.417-421.
[24] Zhang S,Tay Y,Yao L,et al.Next Item Recommendation with Self-Attention[OL].https://arxiv.org/abs/1808.06414v2,2018.
[25] Zhou G,Zhu X,Song C,et al.Deep interest network for click-through rate prediction[A].Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining[C].New York:Association for Computing Machinery,2018.1059-1068.
[26] Kang W C,McAuley J.Self-attentive sequential recommendation[A].2018 IEEE International Conference on Data Mining (ICDM)[C].Washington:IEEE Computer Society,2018.197-206.
[27] Goldberg Y,Levy O.word2vec Explained:deriving Mikolov et al.'s negative-sampling word-embedding method[OL].https://arxiv.org/abs/1402.3722v1,2014.
2020, Vol. 48 
