面向任务口语对话系统中不含槽信息话语的端到端对话控制

摘要
参考文献
相关文章 (15)

全文: PDF (1355 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要端到端(end-to-end)模型因其能有效避免传统管道式设计存在的错误传递与累积问题,成为了近年来口语对话系统(spoken dialogue system, SDS)的研究热点。在面向任务SDS的end-to-end对话控制中,处理携带任务领域语义信息(槽信息)的话语可以结合命名实体识别、数据库查询结果等语义特征,而不含槽信息的话语,由于缺乏领域语义信息以及表达多样,其有效对话控制仍然是一个挑战。该文提出一种融合“显式”话语特征和“隐式”上下文信息的end-to-end混合编码网络用于处理不含槽信息话语。具体地,在应用卷积神经网络(convolutional neural network, CNN)对“显式”话语序列提取得到的特征表达的基础上,通过构造和捕获对话序列中“隐式”的系统后台上下文信息,进一步丰富了系统动作分类模型的特征表达。在限定领域面向中文任务SDS中的评估结果表明,与传统的管道式SDS和经典的end-to-end SDS相比,该文的方案在不含槽信息话语的单回合处理以及对话段整体性能上都得到了显著提升。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	黄锵嘉
	黄沛杰
	李杨辉
	杜泽峰

关键词 ：口语对话系统, 端到端, 卷积神经网络, 显式话语特征, 隐式上下文

Abstract：The end-to-end dialogue control of utterances without slot values is a challenging issue. This paper proposes an end-to-end hybrid coding network that combines explicit utterance features and implicit context information to handle utterances without slot information. Specifically, on the basis of feature expressions extracted from the "explicit" dialogue sequence by the convolutional neural network (CNN), the system action classification model is further enriched by constructing and capturing the "implicit" background system context information in the dialogue sequence. Experiments on the task-oriented restricted domain Chinese SDS shows that, compared to the existing methods, the proposed method achieves significant improvements in both per-response accuracy and per-dialog accuracy.

Key words： spoken dialogue system end-to-end CNN explicit utterance features implicit context

收稿日期: 2018-09-29

中图分类号:

TP391

基金资助:国家自然科学基金(71472068);国家级大学生创新训练计划项目(201710564154)

作者简介: 黄锵嘉(1994—),学士,主要研究领域为自然语言处理。E-mail:qjHuang1024@163.com
黄沛杰(1980—),通信作者,博士,副教授,主要研究领域为人工智能、自然语言处理、口语对话系统。E-mail:pjhuang@scau.edu.cn
李杨辉(1995—),硕士研究生,主要研究领域为自然语言处理。Email:lyh_liyanghui@163.com

链接本文:

http://jcip.cipsc.org.cn/CN/Y2018/V32/I12/109

[1] Zue V,et al.JUPITER: a telephone-based conversational interface for weather information[J].IEEE Transactions on Speech and Audio Processing,2000,8(1): 85-96.
[2] Bordes A,Weston J.Learning end-to-end goal-oriented dialog[C]//Proceedings of the 5th International Conferenceon Learning Representations(ICLR 2017),2017:1-15.
[3] Williams J D,Asadi K,Zweig G.Hybrid code networks: practical and efcient end-to-end dialog control with supervised and reinforcement learning [C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (ACL 2017),2015: 665-677.
[4] Wen T H,et al.A network-based end-to-end trainable task-oriented dialogue system[C]//Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2017),2017: 438-449.
[5] Huang P J,et al.Ch2R: a Chinese chatter robot for online shopping guide[C]//Proceedings of the 3rd CIPS-SIGHAN Joint Conference on Chinese Language Processing (CLP-2014),2014: 26-34.
[6] Pappu A,Rudnicky A.The structure and generality of spoken route instructions[C]//Proceedings of the 13th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL 2012),2012: 99-107.
[7] 黄寅飞,等.校园导航系统EasyNav的设计与实现[J].中文信息学报,2001,15(4): 35-40.
[8] Reichel C S,et al.Out-of-domain spoken dialogs in the car: a WoZ study[C]//Proceedings of the 15th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL 2014),2014: 12-21.
[9] Vinyals O,Le Q V.A neural conversational model [C]//Proceedings of the ICML Deep Learning Workshop,Lille,France,2015.
[10] Shang L F,Lu Z D,Li H.Neural responding machine for short-text conversation [C]//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics (ACL 2015),2015: 1577-1586.
[11] Serban I V,et al.Hierarchical neural network generative models for movie dialogues [C]//Proceedings of the 13th AAAI Conference on Artificial Intelligence (AAAI-16),3776-3783.
[12] Sukhbaatar S,Szlam A,Weston J.End-to-end memory networks [C]//Proceedings of the 29th Annual Conference on Neural Information Processing Systems (NIPS 2015),2015:2440-2448.
[13] Raux A,et al.Let’s go public! taking a spoken dialog system to the real world[C]//Proceedings of the European Conference on Speech Communication and Technology (EUROSPEECH 2005),2005:885-888.
[14] Young S,et al.POMDP-based statistical spoken dialog systems: A review [J].Proceedings of IEEE,2013,101(5):1160-1179.
[15] Zhao T C,Eskenazi M.Towards end-to-end learning for dialog state tracking and management using deep reinforcement learning[C]//Proceedings of the 17th Annual Meeting of the Special Interest Group on Discourse and Dialogue(SIGDIAL 2016),2016:1-10.
[16] Sutskever I,Vinyals O,Le Q V.Sequence to sequence learning with neural networks [C]//Proceedings of the 28th Annual Conference on Neural Information Processing Systems (NIPS 2014),2014: 3104-3112.
[17] Cho K.,et al.Learning phrase representations using rnn encoder-decoder for statistical machine translation [C]//Proceedings of the 19th Conference on Empirical Methods in Natural Language Processing (EMNLP 2014),2014: 1724-1734.
[18] Kim Y.Convolutional neural networks for sentence classification[C]//Proceedings of the 19th Conference on Empirical Methods in Natural Language Processing (EMNLP 2014),2014: 1746-1751.
[19] Severyn A,Moschitti A.Learning to rank short text pairs with convolutional deep neural networks[C]//Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval,(ACM 2015),2015: 373-382.
[20] Ciresan D C,et al.Flexible,high performance convolutional neural networks for image classification[C]//Proceedings of the 22nd International Joint Conference on Artificial Intelligence (IJCAI 2011),2011:1237-1242.
[21] Wang J D,et al.Dialogue act recognition for Chinese out-of-domain utterances using hybrid CNN-RF[C]//Proceedings of the 20th International Conference on Asian Language Processing (IALP 2016),2016:14-17.
[22] 张伟男,张杨子,刘挺.对话系统评价方法综述[J].中国科学:信息科学,2017,43(8): 954-966.
[23] Yang X S,et al.End-to-end joint learning of natural language understanding and dialogue manager[C]//Proceedings of the 42nd IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP 2017),2017: 5690-5694.
[24] Ritter A,et al.Data-driven response generation in social media[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2011),2011:583-593.
[25] Shang L F,Lu Z D,Li H.Neural responding machine for short-text conversation[C]//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics(ACL 2015),2015:1577-1586.
[26] Lowe R,et al.The ubuntu dialogue corpus: A large dataset for research in unstructured multi-turn dialogue systems[C]//Proceedings of the 16th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL 2015),2015: 285-294.
[27] Jurcicek F,et al.Real user evaluation of spoken dialogue systems using Amazon Mechanical Turk [C]//Proceedings of the 12th Annual Conference of the International Speech Communication Association (INTERSPEECH 2011),2011: 3061-3064.