任務型對話系統在近幾年十分熱門，提供特定項目的服務，例如：預定餐廳、機票預訂。對話系統不斷有新框架或新方法來被提出，但卻鮮少有人注意完整性和模組的缺陷，又或是只專注對話中的細部元件，大量產出各種框架在數據指標間競賽，但卻忽略對話系統最終目的是要用於線上部屬，本文會探討現有對話系統的方法與適用情境，並使用現有框架Hybrid Code Networks(HCN)與語料資料集dialogue state tracking challenge 6(DSTC6)訓練來完成對話的進行，實踐完整的餐廳對話推薦系統並加入個人化與採用K-means分群演算法解決新使用者。
我們使用深層自編碼器(Deep AutoEncoders)方法來解決推薦所遇上的稀疏問題，為了驗證此方法，我們使用Yelp資料集，該資料有顧客對店家的評分，實驗結果顯示我們的方法優於先前的AutoEncoders，上個段落提到的模組缺陷，我們所使用HCN框架中存在一個對話狀態追蹤(Dialogue State Tracking)缺陷，我們會使用Neural Belief Tracker來解決，最終提出一個整合與改善的餐廳個人化推薦對話系統。

The task-based dialogue system is a popular issue in recent years, which provides services for specific task, such as booking restaurants and airline reservations. There are lots of new frameworks or new methods for dialogue systems; however, people pay little attention to the integrity and module defects. Instead, people only specialize in the detailed modules in the dialogue, and produce a large number of frameworks for the competition. In this regard, people tend to ignore the dialogue system's goal of deploying customer-facing dialog systems. We leverage the Hybrid Code Networks (HCN) framework trained on dialog state tracking challenge 6 (DSTC6). Furthermore, we implement restaurant recommender system with personalization and adopt K-means clustering algorithm to solve new user problem.
We use deep AutoEncoders to tackle with the problem of data sparse for recommender system. To validate the result, we use the Yelp dataset, which includes the customer's rating on the store. The result shows that our method outperforms AutoEncoders in term of benchmark indicators we adopted. And lastly, we use Neural Belief Tracker to solve the problem of dialog state tracking in HCN framework.

目錄
第一章 緒論 1
1.1 背景與動機 1
1.2 研究目的 2
第二章 文獻探討 4
2.1 對話系統分類 4
2.1.1 非任務導向(Non-Task-Oriented) 4
2.1.2 任務導向(Task-Oriented) 5
2.2 任務導向方法 6
2.2.1 流水線型(Pipeline) 7
2.2.2 端到端(End-to-End) 8
2.3 推薦系統方法 8
2.3.1 傳統推薦方法 8
2.3.2 深度學習(Deep Learning) 9
2.3.3 Autoencoders 9
2.4 語料種類 11
2.4.1 人人對話與人機對話(Human-Human Vs. Human-Machine) 11
2.4.2 人機對話語料 11
第三章 研究方法 12
3.1 系統架構 12
3.2 主要架構Hybrid Code Networks(HCN) 13
3.3 Deep Autoencoders(DAE) 14
3.4 個人化(Personalization) 17
3.5 Knowledge-based Recommender 20
3.6 Neural Belief Tracker 22
第四章 實驗結果與評估 24
4.1 資料集 24
4.1.1 Dialog State Tracking Challenge 6 (DSTC6) 24
4.1.2 Yelp資料集 28
4.1.3 Yelp資料集處理 28
4.2 HCN框架評估 29
4.3 DAE 模型測試 30
4.3.1 DAE 參數評估 31
4.3.2 DAE 方法評估 36
4.4 KnowBase Recommender評估 37
4.5 Personalization評估 38
4.6 Neural Belief Tracker評估 39
4.7 實驗結果 40
4.7.1 HCN與DAE 40
4.7.2 HCN與KnowBaseDAE 41
4.7.3 Personalization 42
4.7.4 Neural Belief Tracker 43
4.8 Summary 45
第五章 結論與未來研究 46
5.1 結論 46
5.2 未來研究 46

[1] H. Chen, X. Liu, D. Yin, and J. Tang, "A survey on dialogue systems: Recent advances and new frontiers," ACM SIGKDD Explorations Newsletter, vol. 19, pp. 25-35, 2017.
[2] J. Gao, M. Galley, and L. Li, "Neural approaches to conversational AI," in The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, 2018, pp. 1371-1374.
[3] J. D. Williams, K. Asadi, and G. Zweig, "Hybrid code networks: practical and efficient end-to-end dialog control with supervised and reinforcement learning," in Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2017, pp. 665-677.
[4] A. Bordes, Y.-L. Boureau, and J. Weston, "Learning end-to-end goal-oriented dialog," arXiv preprint arXiv:1605.07683, 2016.
[5] W. Liang and M. Yang, "Hierarchical Hybrid Code Networks for Task-Oriented Dialogue," in Intelligent Computing Theories and Application, Cham, 2018, pp. 194-204.
[6] S. Wu, W. Ren, C. Yu, G. Chen, D. Zhang, and J. Zhu, "Personal recommendation using deep recurrent neural networks in NetEase," in 2016 IEEE 32nd International Conference on Data Engineering (ICDE), 2016, pp. 1218-1229.
[7] A. Gogna and A. Majumdar, "A Comprehensive Recommender System Model: Improving Accuracy for Both Warm and Cold Start Users," IEEE Access, vol. 3, pp. 2803-2813, 2015.
[8] S. Deng, L. Huang, G. Xu, X. Wu, and Z. Wu, "On Deep Learning for Trust-Aware Recommendations in Social Networks," IEEE Transactions on Neural Networks and Learning Systems, vol. 28, pp. 1164-1177, 2017.
[9] S. Zhang, L. Yao, A. Sun, and Y. Tay, "Deep Learning Based Recommender System: A Survey and New Perspectives," ACM Comput. Surv., vol. 52, pp. 1-38, 2019.
[10] J. Ham, S. Lim, K.-H. Lee, and K.-E. Kim, "Extensions to hybrid code networks for FAIR dialog dataset," Computer Speech & Language, vol. 53, pp. 80-91, 2019.
[11] M. Elahi, F. Ricci, and N. Rubens, "A survey of active learning in collaborative filtering recommender systems," Computer Science Review, vol. 20, pp. 29-50, 2016.
[12] S. Lloyd, "Least squares quantization in PCM," IEEE Transactions on Information Theory, vol. 28, pp. 129-137, 1982.
[13] Y. Sun and Y. Zhang, "Conversational Recommender System," arXiv preprint arXiv:1806.03277, 2018.
[14] K. Christakopoulou, F. Radlinski, and K. Hofmann, "Towards Conversational Recommender Systems," presented at the Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA, 2016.
[15] M. Henderson, B. Thomson, and S. Young, "Robust dialog state tracking using delexicalised recurrent neural networks and unsupervised adaptation," in 2014 IEEE Spoken Language Technology Workshop (SLT), 2014, pp. 360-365.
[16] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Dropout: a simple way to prevent neural networks from overfitting," J. Mach. Learn. Res., vol. 15, pp. 1929-1958, 2014.
[17] N. Mrkšić, D. O. Séaghdha, T.-H. Wen, B. Thomson, and S. Young, "Neural belief tracker: Data-driven dialogue state tracking," arXiv preprint arXiv:1606.03777, 2016.
[18] V. Zhong, C. Xiong, and R. Socher, "Global-Locally Self-Attentive Encoder for Dialogue State Tracking," in Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, Australia, 2018, pp. 1458-1467.
[19] J. Perez and F. Liu, "Dialog state tracking, a machine reading approach using Memory Network," in Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics, 2017, pp. 305-314.
[20] P. Xu and Q. Hu, "An End-to-end Approach for Handling Unknown Slot Values in Dialogue State Tracking," in Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 2018, pp. 1448-1457.
[21] L. Ren, K. Xie, L. Chen, and K. Yu, "Towards Universal Dialogue State Tracking," in Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, 2018, pp. 2780-2786.
[22] Y. Li, H. Su, X. Shen, W. Li, Z. Cao, and S. Niu, "Dailydialog: A manually labelled multi-turn dialogue dataset," arXiv preprint arXiv:1710.03957, 2017.
[23] L. Shang, Z. Lu, and H. Li, "Neural responding machine for short-text conversation," arXiv preprint arXiv:1503.02364, 2015.
[24] H. Wang, Z. Lu, H. Li, and E. Chen, "A dataset for research on short-text conversations," in Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, 2013, pp. 935-945.
[25] L. Deng and Y. Liu, Deep Learning in Natural Language Processing: Springer Publishing Company, Incorporated, 2018.
[26] X. Li, Z. C. Lipton, B. Dhingra, L. Li, J. Gao, and Y.-N. Chen, "A user simulator for task-completion dialogues," arXiv preprint arXiv:1612.05688, 2016.
[27] S. Young, "Using POMDPS for dialog management," in 2006 IEEE Spoken Language Technology Workshop, 2006, pp. 8-13.
[28] I. V. Serban, R. Lowe, P. Henderson, L. Charlin, and J. Pineau, "A survey of available corpora for building data-driven dialogue systems," arXiv preprint arXiv:1512.05742, 2015.
[29] X. Li, Y.-N. Chen, L. Li, J. Gao, and A. Celikyilmaz, "End-to-end task-completion neural dialogue systems," arXiv preprint arXiv:1703.01008, 2017.
[30] B. Liu, G. Tur, D. Hakkani-Tur, P. Shah, and L. Heck, "End-to-end optimization of task-oriented dialogue model with deep reinforcement learning," in NIPS Workshop on Conversational AI, 2017.
[31] T. Zhao and M. Eskenazi, "Towards end-to-end learning for dialog state tracking and management using deep reinforcement learning," in Proceedings of the 17th Annual Meeting of the Special Interest Group on Discourse and Dialogue, 2016, pp. 1-10.
[32] T.-H. Wen, D. Vandyke, N. Mrksic, M. Gasic, L. M. Rojas-Barahona, P.-H. Su, et al., "A Network-based End-to-End Trainable Task-oriented Dialogue System," in Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics, 2017, pp. 438-449.
[33] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, vol. 521, p. 436, 2015.
[34] J. S. Breese, D. Heckerman, and C. Kadie, "Empirical analysis of predictive algorithms for collaborative filtering," in Proceedings of the Fourteenth Conference on Uncertainty in Artificial Intelligence, 1998, pp. 43-52.
[35] J. L. Herlocker, J. A. Konstan, L. G. Terveen, and J. T. Riedl, "Evaluating collaborative filtering recommender systems," ACM Transactions on Information Systems (TOIS), vol. 22, pp. 5-53, 2004.
[36] S. Okura, Y. Tagami, S. Ono, and A. Tajima, "Embedding-based News Recommendation for Millions of Users," presented at the Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Halifax, NS, Canada, 2017.
[37] H.-T. Cheng, L. Koc, J. Harmsen, T. Shaked, T. Chandra, H. Aradhye, et al., "Wide & Deep Learning for Recommender Systems," presented at the Proceedings of the 1st Workshop on Deep Learning for Recommender Systems, Boston, MA, USA, 2016.
[38] C. A. Gomez-Uribe and N. Hunt, "The netflix recommender system: Algorithms, business value, and innovation," ACM Transactions on Management Information Systems (TMIS), vol. 6, p. 13, 2016.
[39] P. Covington, J. Adams, and E. Sargin, "Deep Neural Networks for YouTube Recommendations," in Proceedings of the 10th ACM Conference on Recommender Systems, Boston, Massachusetts, USA, 2016, pp. 191-198.
[40] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," nature, vol. 521, p. 436, 2015.
[41] C.-Y. Wu, A. Ahmed, A. Beutel, A. J. Smola, and H. Jing, "Recurrent Recommender Networks," presented at the Proceedings of the Tenth ACM International Conference on Web Search and Data Mining, Cambridge, United Kingdom, 2017.
[42] S. T. Jishan and Y. Wang, "Audience Activity Recommendation Using Stacked-LSTM Based Sequence Learning," presented at the Proceedings of the 9th International Conference on Machine Learning and Computing, Singapore, Singapore, 2017.
[43] S. Sedhain, A. K. Menon, S. Sanner, and L. Xie, "Autorec: Autoencoders meet collaborative filtering," in Proceedings of the 24th International Conference on World Wide Web, 2015, pp. 111-112.
[44] O. Kuchaiev and B. Ginsburg, "Training deep autoencoders for collaborative filtering," arXiv preprint arXiv:1708.01715, 2017.
[45] P. Vincent, H. Larochelle, Y. Bengio, and P.-A. Manzagol, "Extracting and composing robust features with denoising autoencoders," presented at the Proceedings of the 25th international conference on Machine learning, Helsinki, Finland, 2008.
[46] G. E. Hinton and R. S. Zemel, "Autoencoders, minimum description length and Helmholtz free energy," presented at the Proceedings of the 6th International Conference on Neural Information Processing Systems, Denver, Colorado, 1993.
[47] M. Cogswell, F. Ahmed, R. Girshick, L. Zitnick, and D. Batra, "Reducing Overfitting in Deep Networks by Decorrelating Representations," CoRR, vol. abs/1511.06068, 2015.
[48] J. D. Williams and S. Young, "Partially observable Markov decision processes for spoken dialog systems," Computer Speech & Language, vol. 21, pp. 393-422, 2007.
[49] J. Williams, A. Raux, D. Ramachandran, and A. Black, "The Dialog State Tracking Challenge," in Proceedings of the SIGDIAL 2013 Conference, Metz, France, 2013, pp. 404-413.
[50] M. Henderson, B. Thomson, and J. D. Williams, "The Second Dialog State Tracking Challenge," in Proceedings of the 15th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL), Philadelphia, PA, U.S.A., 2014, pp. 263-272.
[51] S. Kim, L. F. D’Haro, R. E. Banchs, J. D. Williams, and M. Henderson, "The fourth dialog state tracking challenge," in Dialogues with Social Robots, ed: Springer, 2017, pp. 435-449.
[52] Y. Boureau, A. Bordes, and J. Perez, "Dialog state tracking challenge 6 end-to-end goal-oriented dialog track," Tech. Rep., Tech. Rep, 2017.
[53] Y. Yang and T. M. Hospedales, "Deep neural networks for sketch recognition," arXiv preprint arXiv:1501.07873, vol. 1, p. 3, 2015.
[54] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Dropout: a simple way to prevent neural networks from overfitting," The journal of machine learning research, vol. 15, pp. 1929-1958, 2014.
[55] C. C. Aggarwal, Recommender Systems: The Textbook: Springer Publishing Company, Incorporated, 2016.
[56] C. Févotte and J. Idier, "Algorithms for Nonnegative Matrix Factorization with the β-Divergence," Neural Computation, vol. 23, pp. 2421-2456, 2011.
[57] T. Mikolov, I. Sutskever, K. Chen, G. Corrado, and J. Dean, "Distributed representations of words and phrases and their compositionality," presented at the Proceedings of the 26th International Conference on Neural Information Processing Systems - Volume 2, Lake Tahoe, Nevada, 2013.
[58] M. D. Zeiler, "ADADELTA: an adaptive learning rate method," arXiv preprint arXiv:1212.5701, 2012.
[59] G. Bai and Z. Zhan, "Application of Deep Autoencoders in Commerce Recommendation," in Advances in Computer Communication and Computational Sciences, ed: Springer, 2019, pp. 235-242.
[60] C. K. Joshi, F. Mi, and B. Faltings, "Personalization in Goal-oriented Dialog," arXiv preprint arXiv:1706.07503, 2017.
[61] S. Zhang, E. Dinan, J. Urbanek, A. Szlam, D. Kiela, and J. Weston, "Personalizing Dialogue Agents: I have a dog, do you have pets too?," arXiv preprint arXiv:1801.07243, 2018.
[62] G. Linden, B. Smith, and J. York, "Amazon.com recommendations: item-to-item collaborative filtering," IEEE Internet Computing, vol. 7, pp. 76-80, 2003.
[63] F. Ricci, L. Rokach, and B. Shapira, "Introduction to recommender systems handbook," in Recommender systems handbook, ed: Springer, 2011, pp. 1-35.
[64] U. Shardanand and P. Maes, "Social information filtering: algorithms for automating “word of mouth&rdquo," presented at the Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Denver, Colorado, USA, 1995.
[65] Y. Deldjoo, M. F. Dacrema, M. G. Constantin, H. Eghbal-Zadeh, S. Cereda, M. Schedl, et al., "Movie genome: alleviating new item cold start in movie recommendation," User Modeling and User-Adapted Interaction, vol. 29, pp. 291-343, 2019.
[66] C. Hori, J. Perez, R. Higashinaka, T. Hori, Y. L. Boureau, M. Inaba, et al., "Overview of the sixth dialog system technology challenge: DSTC6," Computer Speech & Language, vol. 55, pp. 1-25, 2019.
[67] X. He, H. Zhang, M.-Y. Kan, and T.-S. Chua, "Fast matrix factorization for online recommendation with implicit feedback," in Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2016, pp. 549-558.
[68] X. He, Z. He, X. Du, and T.-S. Chua, "Adversarial personalized ranking for recommendation," in The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, 2018, pp. 355-364.
[69] C. Yang, L. Bai, C. Zhang, Q. Yuan, and J. Han, "Bridging collaborative filtering and semi-supervised learning: A neural approach for poi recommendation," in Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2017, pp. 1245-1254.
[70] W. Sun, F. Su, and L. Wang, "Improving deep neural networks with multi-layer maxout networks and a novel initialization method," Neurocomputing, vol. 278, pp. 34-40, 2018.
[71] X. Glorot and Y. Bengio, "Understanding the difficulty of training deep feedforward neural networks," presented at the Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, Proceedings of Machine Learning Research, 2010.
[72] D.-A. Clevert, T. Unterthiner, and S. Hochreiter, "Fast and accurate deep network learning by exponential linear units (elus)," arXiv preprint arXiv:1511.07289, 2015.
[73] G. Klambauer, T. Unterthiner, A. Mayr, and S. Hochreiter, "Self-normalizing neural networks," presented at the Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, California, USA, 2017.
[74] R. Catherine and W. Cohen, "Transnets: Learning to transform for recommendation," in Proceedings of the Eleventh ACM Conference on Recommender Systems, 2017, pp. 288-296.
[75] L. Zheng, V. Noroozi, and P. S. Yu, "Joint deep modeling of users and items using reviews for recommendation," in Proceedings of the Tenth ACM International Conference on Web Search and Data Mining, 2017, pp. 425-434.
[76] S. Consolvo, L. Arnstein, and B. R. Franza, "User study techniques in the design and evaluation of a ubicomp environment," in International Conference on Ubiquitous Computing, 2002, pp. 73-90.
[77] C. C. Aggarwal, Recommender systems: Springer, 2016.
[78] Z. Zhang, Q. Wang, L. Ruan, and L. Si, "Preference preserving hashing for efficient recommendation," in Proceedings of the 37th international ACM SIGIR Conference on Research & Development in Information Retrieval, 2014, pp. 183-192.
[79] H. Oosterhuis and M. de Rijke, "Ranking for relevance and display preferences in complex presentation layouts," in The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, 2018, pp. 845-854.