推薦系統可區分三種類型，包跨矩陣完全類型與短序列類型與長序列型。本論文主要探討短序列類型與長序列型。這兩種類型的推薦系統主要差別於時間的長短。這篇論文有三種應用場景。第一個場景是想要預測淺在的失眠患者。我們收集使用者的睡眠資料，並且建立模型，模型的輸出是所有使用者的睡眠效率排名，其中睡眠效率越低的會被排到較前面的排名。相反的，如果睡眠效率越高則會被排名到較後面的排名。此外，這類型的推薦系統是屬於短序列的推薦係統。第二個應用場景是公平性的推薦系統。公平性是非常重要的。舉例來說，如果我們所訓練的資料有包含性別種族等等的人的屬性並且比例嚴重的不均勻，這會造成預測上的不穩定與無法達到良好的預測，為了讓預測更準確，我們有在模型裡面增加多工作學習的技術， 把知識圖譜應用再我們的模型當中。此外，這次所使用的推薦系統是長序列類型的推薦系統，更能符合生活的應用場景。第三個應用場景是對話性的推薦系統。如果讓推薦系統具有對話性的功能，這樣更能夠讓推薦系統能在第一線與使用者進行交流，並且獲得更即時的訊息。在推薦系統的部分，我們為了與對話性的部分進行連接，我們需要把遷移是學習的技術加入到我們的推薦系統，這才能達到即時的更新使用者與商品的資訊。此外，這個推薦系統是屬於長序列的推薦系統。
關鍵字: 遷移學習, 推薦系統, 個人化學習, 長序列學習, 公平性

Recommendation System can distinguish three type, including matrix completion based, ses- sion based and sequential recommendation. The paper focus on session based and sequential based. The main difference of them is the length of time. Session based belongs to short term model, sequential based belongs to long term model. We adopt different recommen- dation type depended on applied scenario. One of scenario is forecast of latent insomnia patient. We collect user sleep data by wearable device and build a model to generate user’s sleep efficient rank list. We can forecast latent high risk insomnia patient based on the rank- ing list. We propose a model, Pairwise Learning-based Ranking Generation (PLRG), to rank users with high insomnia potential in the next day. Moreover, we consider fairness and con- versational system in the paper. Fairness-aware recommendation mitigates a variety of al- gorithmic bias in the learning of user preferences. It aims at bringing a marriage between Sequential recommendation and algorithmic fairness. We propose a novel fairness-aware se- quential recommendation task, in which a new metric, interaction fairness, is defined to esti- mate how recommended items are fairly interacted by users with different protected attribute groups. We propose a multi-task learning based deep end-to-end model, FairSR. Conversa- tional recommendation system have two main core, including conversational component and recommendation component. conversational component make conversation with user and it will receive some user latent habit. recommendation component receive user’s habit and then make more precise recommendation. We propose a inductive learning based seep end to end model, CPRS. Extensive experiments conducted on three datasets show FairSR, CPRS can outperform state-of-the-art SR models in recommendation performance. In addition, the recommended items by FairSR also exhibit promising interaction fairness.
Keyword: Inductive learning, Transfer learning, Recommendation System, Personalized Learning, Fairness

[1] Himan Abdollahpouri. Popularity bias in ranking and recommendation. In Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society, AIES ’19, pages 529– 530, 2019.
[2] Himan Abdollahpouri, Masoud Mansoury, Robin Burke, and Bamshad Mobasher. The impact of popularity bias on fairness and calibration in recommendation, 2019.
[3] Rajendra Acharya, Oliver Faust, N Kannathal, TjiLeng Chua, and Swamy Laxmi- narayan. Non-linear analysis of eeg signals at various sleep stages. Elsevier Computer Methods and Programs in Biomedicine, 80(1):37–45, 2005.
[4] Dzmitry Bahdanau, Kyunghyun Cho, and Yoshua Bengio. Neural machine translation by jointly learning to align and translate, 2014.
[5] Alex Beutel, Jilin Chen, Tulsee Doshi, Hai Qian, Li Wei, Yi Wu, Lukasz Heldt, Zhe Zhao, Lichan Hong, Ed H. Chi, and et al. Fairness in recommendation ranking through pairwise comparisons. In Proceedings of the 25th ACM SIGKDD International Con- ference on Knowledge Discovery & Data Mining, KDD ’19, pages 2212–2220, 2019.
[6] Tessa F Blanken, Jeroen S Benjamins, Denny Borsboom, Jeroen K Vermunt, Casey Paquola, Jennifer Ramautar, Kim Dekker, Diederick Stoffers, Rick Wassing, Yishul Wei, et al. Insomnia disorder subtypes derived from life history and traits of affect and personality. Elsevier Lancet Psychiatry, 2019.
[7] Avishek Bose and William Hamilton. Compositional fairness constraints for graph em- beddings. In Proceedings of the 36th International Conference on Machine Learning, volume 97 of ICML ’19, pages 715–724, 2019.
[8] Joan Bruna, Wojciech Zaremba, Arthur Szlam, and Yann LeCun. Spectral networks and locally connected networks on graphs, 2013.
[9] Chris J.C. Burges. From ranknet to lambdarank to lambdamart: An overview. Technical Report, Microsoft Research, 2010.
[10] Danilo Bzdok and Andreas Meyer-Lindenberg. Machine learning for precision psychi- atry: Opportunites and challenges. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 2017.
[11] Yixin Cao, Xiang Wang, Xiangnan He, Zikun Hu, and Tat-Seng Chua. Unifying knowl- edge graph learning and recommendation: Towards a better understanding of user pref- erences. In The World Wide Web Conference, WWW ’19, pages 151–161, 2019.
[12] Tianqi Chen and Carlos Guestrin. Xgboost: A scalable tree boosting system. In proc. of the ACM SIGKDD, 2016.
[13] Edward Choi, Mohammad Taha Bahadori, Elizabeth Searles, Catherine Coffey, Michael Thompson, James Bost, Javier Tejedor-Sojo, and Jimeng Sun. Multi-layer representation learning for medical concepts. In proc. of the ACM SIGKDD, 2016.
[14] Yann N. Dauphin, Angela Fan, Michael Auli, and David Grangier. Language modeling with gated convolutional networks. In Proceedings of the 34th International Conference on Machine Learning - Volume 70, ICML ’17, pages 933–941, 2017.
[15] Michaël Defferrard, Xavier Bresson, and Pierre Vandergheynst. Convolutional neural networks on graphs with fast localized spectral filtering, 2016.
[16] Hui Ding, Goce Trajcevski, Peter Scheuermann, Xiaoyue Wang, and Eamonn Keogh. Querying and mining of time series data: experimental comparison of representations and distance measures. In proc. of the VLDB, 2008.
[17] Michael D. Ekstrand, Mucun Tian, Ion Madrazo Azpiazu, Jennifer D. Ekstrand, Oghen- emaro Anuyah, David McNeill, and Maria Soledad Pera. All the cool kids, how do they fit in?: Popularity and demographic biases in recommender evaluation and effective- ness. In Proceedings of the 1st Conference on Fairness, Accountability and Trans- parency, volume 81, pages 172–186, 2018.
[18] Kelly R Evenson, Michelle M Goto, and Robert D Furberg. Systematic review of the va- lidity and reliability of consumer-wearable activity trackers. IJBNPA, 12(1):159, 2015.
[19] Golnoosh Farnadi, Pigi Kouki, Spencer K. Thompson, Sriram Srinivasan, and Lise Getoor. A fairness-aware hybrid recommender system. In Proceedings of 2nd FATREC Workshop: Responsible Recommendation, 2018.
[20] Brisa S Fernandes, Leanne M Williams, Johann Steiner, Marion Leboyer, André F Car- valho, and Michael Berk. The new field of ‘precision psychiatry’. BMC medicine, 15(1):80, 2017.
[21] Sahin Cem Geyik, Stuart Ambler, and Krishnaram Kenthapadi. Fairness-aware rank- ing in search and recommendation systems with application to linkedin talent search. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Dis- covery & Data Mining, KDD ’19, pages 2221–2231, 2019.
[22] Michele Gori, Gabriele Monfardini, and Franco Scarselli. A new model for learning in graph domains. Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005., 2:729–734 vol. 2, 2005.
[23] Gregory Gutin and Abraham P. Punnen. The traveling salesman problem and its varia- tions. Combinatorial Optimization, 2007.
[24] William L. Hamilton, Rex Ying, and Jure Leskovec. Inductive representation learning on large graphs, 2017.
[25] Xiangnan He, Hanwang Zhang, Min-Yen Kan, and Tat-Seng Chua. Fast matrix factor- ization for online recommendation with implicit feedback. In Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’16, pages 549–558, 2016.
[26] Balázs Hidasi and Alexandros Karatzoglou. Recurrent neural networks with top-k gains for session-based recommendations. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management, CIKM ’18, pages 843–852, 2018.
[27] Balazs Hidasi, Alexandros Karatzoglou, Linas Baltrunas, and Domonkos Tikk. Session- based recommendations with recurrent neural networks. In Proceedings of International Conference on Learning Representations, ICLR ’16, 2016.
[28] Wahyu Hidayat, Toufan D Tambunan, and Reza Budiawan. Empowering wearable sensor generated data to predict changes in individual’s sleep quality. In proc. of the IEEE ICoICT, 2018.
[29] Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory. Neural computa- tion, 9(8):1735–1780, 1997.
[30] Jyun-Yu Jiang, Cheng-Te Li, Yian Chen, and Wei Wang. Identifying users behind shared accounts in online streaming services. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, SIGIR ’18, pages 65–74, 2018.
[31] David A. Kalmbach, J. Todd Arnedt, Vivek Pillai, and Christopher L. Drake. Identifying at-risk individuals for insomnia using the ford insomnia response to stress test. Sleep, 39(2):449–456, 2016.
[32] W. Kang and J. McAuley. Self-attentive sequential recommendation. In 2018 IEEE International Conference on Data Mining (ICDM), pages 197–206, 2018.
[33] Uri Kartoun, Rahul Aggarwal, Andrew Beam, Jennifer K. Pai, Arnaub Chatterjee, Tim- othy P. Fitzgerald, Isaac Kohane, and Stanley Shaw. Development of an algorithm to identify patients with physician-documented insomnia. Scientific Reports, 8:Article Number: 7862, 2018.
[34] Diederik P. Kingma and Jimmy Lei Ba. Adam: A method for stochastic optimization. In Proceedings of International Conference on Learning Representations, ICLR ’15, 2015.
[35] Thomas N. Kipf and Max Welling. Semi-supervised classification with graph convolu- tional networks, 2016.
[36] Thomas N. Kipf and Max Welling. Semi-supervised classification with graph convolu- tional networks, 2016.
[37] Srijan Kumar, Xikun Zhang, and Jure Leskovec. Predicting dynamic embedding tra- jectory in temporal interaction networks. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD ’19, 2019.
[38] Yujia Li, Daniel Tarlow, Marc Brockschmidt, and Richard Zemel. Gated graph sequence neural networks, 2015.
[39] Chen Ma, Peng Kang, and Xue Liu. Hierarchical gating networks for sequential rec- ommendation. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD ’19, pages 825–833, 2019.
[40] Chen Ma, Liheng Ma, Yingxue Zhang, Jianing Sun, Xue Liu, and Mark Coates. Mem- ory augmented graph neural networks for sequential recommendation, 2019.
[41] Weizhi Ma, Min Zhang, Yue Cao, Woojeong Jin, Chenyang Wang, Yiqun Liu, Shaop- ing Ma, and Xiang Ren. Jointly learning explainable rules for recommendation with knowledge graph. In The World Wide Web Conference, WWW ’19, pages 1210–1221, 2019.
[42] Ninareh Mehrabi, Fred Morstatter, Nripsuta Saxena, Kristina Lerman, and Aram Gal- styan. A survey on bias and fairness in machine learning, 2019.
[43] Hannah Morphy, Kate M Dunn, Martyn Lewis, Helen F Boardman, and Peter R Croft. Epidemiology of insomnia: a longitudinal study in a uk population. Sleep, 30(3):274– 280, 2007.
[44] Tien T. Nguyen, Pik-Mai Hui, F. Maxwell Harper, Loren Terveen, and Joseph A. Kon- stan. Exploring the filter bubble: The effect of using recommender systems on content diversity. In Proceedings of the 23rd International Conference on World Wide Web, WWW ’14, pages 677–686, 2014.
[45] Sarun Paisarnsrisomsuk, Michael Sokolovsky, Francisco Guerrero, Carolina Ruiz, and Sergio A. Alvarez. Deep sleep: Convolutional neural networks for predictive modeling of human sleep time-signals. In KDD Deep Learning Day, 2018.
[46] Eli Pariser. The Filter Bubble: What The Internet Is Hiding From You. Penguin Books Limited, 2011.
[47] Bo Peng, Zhiyun Ren, Srinivasan Parthasarathy, and Xia Ning. Ham: Hybrid associa- tions model with pooling for sequential recommendation, 2020.
[48] Tahleen Rahman, Bartlomiej Surma, Michael Backes, and Yang Zhang. Fairwalk: To- wards fair graph embedding. In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI-19, pages 3289–3295, 2019.
[49] Bashir Rastegarpanah, Krishna P. Gummadi, and Mark Crovella. Fighting fire with fire: Using antidote data to improve polarization and fairness of recommender systems. In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining, WSDM ’19, pages 231–239, 2019.
[50] Steffen Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars Schmidt-Thieme. Bpr: Bayesian personalized ranking from implicit feedback. In Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence, UAI ’09, pages 452– 461, 2009.
[51] Aarti Sathyanarayana, Jaideep Srivastava, and Luis Fernandez-Luque. The science of sweet dreams: predicting sleep efficiency from wearable device data. Computer, 50(3):30–38, 2017.
[52] F. Scarselli, M. Gori, A. C. Tsoi, M. Hagenbuchner, and G. Monfardini. The graph neural network model. IEEE Transactions on Neural Networks, 20(1):61–80, 2009.
[53] Michael Schlichtkrull, Thomas N. Kipf, Peter Bloem, Rianne van den Berg, Ivan Titov, and Max Welling. Modeling relational data with graph convolutional networks, 2017.
[54] Chuan Shi, Yitong Li, Jiawei Zhang, Yizhou Sun, and Philip S. Yu. A survey of hetero- geneous information network analysis, 2015.
[55] Jiaxi Tang and Ke Wang. Personalized top-n sequential recommendation via convolu- tional sequence embedding. In Proceedings of the Eleventh ACM International Con- ference on Web Search and Data Mining, WSDM ’18, pages 565–573, 2018.
[56] Rianne van den Berg, Thomas N. Kipf, and Max Welling. Graph convolutional matrix completion, 2017.
[57] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need, 2017.
[58] Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, and Yoshua Bengio. Graph attention networks. In Proceedings of International Con- ference on Learning Representations, ICLR ’18, 2018.
[59] Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, and Yoshua Bengio. Graph attention networks, 2017.
[60] Hongwei Wang, Fuzheng Zhang, Jialin Wang, Miao Zhao, Wenjie Li, Xing Xie, and Minyi Guo. Ripplenet: Propagating user preferences on the knowledge graph for rec- ommender systems. In Proceedings of the 27th ACM International Conference on In- formation and Knowledge Management, CIKM ’18, pages 417–426, 2018.
[61] Hongwei Wang, Fuzheng Zhang, Miao Zhao, Wenjie Li, Xing Xie, and Minyi Guo. Multi-task feature learning for knowledge graph enhanced recommendation. In The World Wide Web Conference, WWW ’19, pages 2000–2010, 2019.
[62] Q. Wang, Z. Mao, B. Wang, and L. Guo. Knowledge graph embedding: A survey of approaches and applications. IEEE Transactions on Knowledge and Data Engineering, 29(12):2724–2743, 2017.
[63] Xiang Wang, Xiangnan He, Yixin Cao, Meng Liu, and Tat-Seng Chua. Kgat: Knowl- edge graph attention network for recommendation. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD ’19, pages 950–958, 2019.
[64] Xiang Wang, Xiangnan He, Meng Wang, Fuli Feng, and Tat-Seng Chua. Neural graph collaborative filtering. In Proceedings of the 42nd International ACM SIGIR Confer- ence on Research and Development in Information Retrieval, SIGIR ’19, pages 165– 174, 2019.
[65] Xiang Wang, Dingxian Wang, Canran Xu, Xiangnan He, Yixin Cao, and Tat-Seng Chua. Multi-order attentive ranking model for sequential recommendation. In Proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence, AAAI-19, pages 5329– 5336, 2019.
[66] Xiao Wang, Houye Ji, Chuan Shi, Bai Wang, Peng Cui, P. Yu, and Yanfang Ye. Het- erogeneous graph attention network, 2019.
[67] Chao-Yuan Wu, Amr Ahmed, Alex Beutel, Alexander J. Smola, and How Jing. Recur- rent recommender networks. In Proceedings of the Tenth ACM International Confer- ence on Web Search and Data Mining, WSDM ’17, page 495–503, 2017.
[68] Sirui Yao and Bert Huang. Beyond parity: Fairness objectives for collaborative filtering. In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS ’17, pages 2925–2934, 2017.
[69] Lu Yu, Chuxu Zhang, Shangsong Liang, and Xiangliang Zhang. Multi-order attentive ranking model for sequential recommendation. In Proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence, AAAI-19, pages 5709–5716, 2019.
[70] Fajie Yuan, Alexandros Karatzoglou, Ioannis Arapakis, Joemon M. Jose, and Xiangnan He. A simple convolutional generative network for next item recommendation. In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining, WSDM ’19, pages 582–590, 2019.
[71] Muhan Zhang and Yixin Chen. Inductive matrix completion based on graph neural networks, 2019.
[72] Shuai Zhang, Lina Yao, Aixin Sun, and Yi Tay. Deep learning based recommender system: A survey and new perspectives. ACM Comput. Surv., 52(1), 2019.
[73] Yang Zhang, Mathias Humbert, Tahleen Rahman, Cheng-Te Li, Jun Pang, and Michael Backes. Tagvisor: A privacy advisor for sharing hashtags. In Proceedings of the 2018 World Wide Web Conference, WWW ’18, pages 287–296, 2018.
[74] Zhi-Hua Zhou and Ji Feng. Deep forest: Towards an alternative to deep neural networks. In proc. of the IJCAI, 2017.
[75] Ziwei Zhu, Xia Hu, and James Caverlee. Fairness-aware tensor-based recommenda- tion. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management, CIKM ’18, pages 1153–1162, 2018.