Journal Home > Volume 5 , issue 1

Dialog State Tracking (DST) aims to extract the current state from the conversation and plays an important role in dialog systems. Existing methods usually predict the value of each slot independently and do not consider the correlations among slots, which will exacerbate the data sparsity problem because of the increased number of candidate values. In this paper, we propose a multi-domain DST model that integrates slot-relevant information. In particular, certain connections may exist among slots in different domains, and their corresponding values can be obtained through explicit or implicit reasoning. Therefore, we use the graph adjacency matrix to determine the correlation between slots, so that the slots can incorporate more slot-value transformer information. Experimental results show that our approach has performed well on the Multi-domain Wizard-Of-Oz (MultiWOZ) 2.0 and MultiWOZ2.1 datasets, demonstrating the effectiveness and necessity of incorporating slot-relevant information.


menu
Abstract
Full text
Outline
About this article

Exploiting More Associations Between Slots for Multi-Domain Dialog State Tracking

Show Author's information Hui BaiYan Yang( )Jie Wang
School of Computing and Artifical Intelligence, Southwest Jiaotong University, Chengdu 611756, China

Abstract

Dialog State Tracking (DST) aims to extract the current state from the conversation and plays an important role in dialog systems. Existing methods usually predict the value of each slot independently and do not consider the correlations among slots, which will exacerbate the data sparsity problem because of the increased number of candidate values. In this paper, we propose a multi-domain DST model that integrates slot-relevant information. In particular, certain connections may exist among slots in different domains, and their corresponding values can be obtained through explicit or implicit reasoning. Therefore, we use the graph adjacency matrix to determine the correlation between slots, so that the slots can incorporate more slot-value transformer information. Experimental results show that our approach has performed well on the Multi-domain Wizard-Of-Oz (MultiWOZ) 2.0 and MultiWOZ2.1 datasets, demonstrating the effectiveness and necessity of incorporating slot-relevant information.

Keywords:

slot-relevant attention, multi-domain dialog state tracking, task-oriented dialog system
Received: 07 July 2021 Accepted: 19 July 2021 Published: 27 December 2021 Issue date: March 2022
References(34)
[1]
F. Li, L. Li, J. Yin, L. Huang, Q. Zhou, N. An, and L. Yu, Machine knowledge and human cognition, Big Data Mining and Analytics, vol. 3, no. 4, pp. 292-299, 2020.
[2]
P. Budzianowski, T. H. Wen, and B. H. Tseng, MultiWOZ-A large-scale multi-domain Wizard-Of-Oz dataset for task-oriented dialogue modelling, arXiv preprint arXiv: 1810.00278, 2018.
[3]
M. Henderson, B. Thomson, and S. Young, Word-based dialog state tracking with recurrent neural networks, in Proc. 5th Annu. Meeting of the Special Interest Group on Discourse and Dialogue, Philadelphia, PA, USA, 2014, pp. 292-299.
[4]
P. Xu and Q. Hu, An end-to-end approach for handling unknown slot values in dialogue state tracking, in Proc. 56th Annu. Meeting of the Association for Computational Linguistics, Melbourne, Australia, 2018, pp. 1448-1457.
[5]
V. Zhong, C. Xiong, and R. Socher, Global-locally self-attentive dialogue state tracker, arXiv preprint arXiv: 1805.09655, 2018.
[6]
L. Ren, K. Xie, L. Chen, and K. Yu, Towards universal dialogue state tracking, in Proc. 2018 Conf. Empirical Methods in Natural Language Processing, Brussels, Belgium, 2018, pp. 2780-2786.
[7]
R. Goel, S. Paul, T. Chung, J. Lecomte, A. Mandal, and D. Hakkani-Tur, Flexible and scalable state tracking framework for goal-oriented dialogue systems, arXiv preprint arXiv: 1811.12891, 2018.
[8]
Z. Wang and O. Lemon, A simple and generic belief tracking mechanism for the dialog state tracking challenge: On the believability of observed information, in Proc. 14th Annu. Meeting of the Special Interest Group on Discourse and Dialogue, Metz, France, 2013, pp. 423-432.
[9]
J. D. Williams, Web-style ranking and SLU combination for dialog state tracking, in Proc. 15th Annu. Meeting of the Special Interest Group on Discourse and Dialogue, Philadelphia, PA, USA, 2014, pp. 282-291.
[10]
J. Perez and F. Liu, Dialog state tracking, a machine reading approach using memory network, in Proc. 15th Conf. European Chapter of the Association for Computational Linguistics, Valencia, Spain, 2017, pp. 305-314.
[11]
L. Zilka and F. Jurcicek, Incremental LSTM-based dialog state tracke, in Proc. 2015 IEEE Workshop on Automatic Speech Recognition and Understanding, Scottsdale, AR, USA, 2015, pp. 757-762.
[12]
T. H. Wen, D. Vandyke, and N. Mrksic, A network-based end-to-end trainable task-oriented dialogue system, in Proc. 15th Conf. European Chapter of the Association for Computational Linguistics, Valencia, Spain, 2017, pp. 438-449.
[13]
N. Mrkšić, D. Ó. Séaghdha, and T. H. Wen, Neural belief tracker: Data-driven dialogue state tracking, in Proc. 55th Annu. Meeting of the Association for Computational Linguistics, Vancouver, Canada, 2017, pp. 1777-1788.
[14]
M. Eric, R. Goel, and S. Paul, Multiwoz 2.1: Multi-domain dialogue state corrections and state tracking baselines, arXiv preprint arXiv: 1907.01669, 2019.
[15]
S. Hochreiter and J. Schmidhuber, Long short-term memory, Neural Comput, vol. 9, no. 8, pp. 1735-1780, 1997.
[16]
H. Lee, J. Lee, and T. Y. Kim, SUMBT: Slot-utterance matching for universal and scalable belief tracking, in Proc. 57th Annu. Meeting of the Association for Computational Linguistics, Florence, Italy, 2019, pp. 5478-5483.
[17]
L. Zhou and K. Small, Multi-domain dialogue state tracking as dynamic knowledge graph enhanced question answering, arXiv preprint arXiv: 1911.06192, 2019.
[18]
L. Chen, B. Lv, C. Wang, S. Zhu, B. Tan, and K. Yu, Schema-guided multi-domain dialogue state tracking with graph attention neural networks, Proc. AAAI Conf. Artificial Intelligence, vol. 34, no. 5, pp. 7521-7528, 2020.
[19]
Y. Shan, Z. Li, J. Zhang, F. Meng, Y. Feng, C. Niu, and J. Zhou, A contextual hierarchical attention network with adaptive objective for dialogue state tracking, arXiv preprint arXiv: 2006.01554, 2020.
[20]
J. Devlin, M. W. Chang, K Lee, and K. Toutanova, BERT: Pre-training of deep bidirectional transformers for language understanding, arXiv preprint arXiv: 1810.04805, 2018.
[21]
S. Gao, A. Sethi, S. Agarwal, T. Chung, and D. Hakkani-Tur, Dialog state tracking: A neural reading comprehension approach, in Proc. 20th Annu. SIGdial Meeting on Discourse and Dialogue, Stockholm, Sweden, 2019, pp. 264-273.
[22]
J. Zhang, K. Hashimoto, C. S. Wu, Y. Wang, S. Y. Philip, R. Socher, and C. Xiong, Find or classify? Dual strategy for slot-value predictions on multi-domain dialog state tracking, in Proc. 9th Conf. Lexical and Computational Semantics, Barcelona, Spain, 2019, pp. 154-167.
[23]
C. S. Wu, A. Madotto, E. Hosseini-Asl, C. Xiong, R. Socher, and P. Fung, Transferable multi-domain state generator for task-oriented dialogue systems, in Proc. 57th Annu. Meeting of the Association for Computational Linguistics, Florence, Italy, 2019, pp. 808-819.
[24]
K. Cho, B. Van Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y. Bengio, Learning phrase representations using RNN encoder-decoder for statistical machine translation, in Proc. 2014 Conf. Empirical Methods in Natural Language Processing, Doha, Qatar, 2014, pp. 1724-1734.
[25]
L. Ren, J. Ni, and J. McAuley, Scalable and accurate dialogue state tracking via hierarchical sequence generation, in Proc. 2019 Conf. Empirical Methods in Natural Language Processing and 9th Int. Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Hong Kong, China, 2019, pp. 1876-1885.
[26]
H. Le, R. Socher, and S. C. Hoi, Non-autoregressive dialog state tracking, arXiv preprint arXiv: 2002.08024, 2020.
[27]
S. Kim, S. Yang, G. Kim, and S. W. Lee, Efficient dialogue state tracking by selectively overwriting memory, in Proc. 58th Annu. Meeting of the Association for Computational Linguistics, .
[28]
S. Zhu, J. Li, L. Chen, and K. Yu, Efficient context and schema fusion networks for multi-domain dialogue state tracking, in Proc. 2020 Conf. Empirical Methods in Natural Language Processing, .
[29]
A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, and I. Polosukhin, Attention is all you need, in Proc. 31th Int. Conf. Neural Information Processing Systems, Long Beach, CA, USA, 2017, pp. 6000-6010.
[30]
C. Banerjee, T. Mukherjee, and E. Pasiliao, Feature representations using the reflected rectified linear unit (RReLU) activation, Big Data Mining and Analytics, vol. 3, no. 2, pp. 102-120, 2020.
[31]
J. L. Ba, J. R. Kiros, and G. E. Hinton, Layer normalization, arXiv preprint arXiv:1607.06450, 2016.
[32]
A. See, P. J. Liu, and C. D. Manning, Get to the point: Summarization with pointer-generator networks, in Proc. 55th Annu. Meeting of the Association for Computational Linguistics, Vancouver, Canada, 2017, pp. 1073-1083.
[33]
B. McCann, N. S. Keskar, C. Xiong, and R. Socher, The natural language decathlon: Multitask learning as question answering, arXiv preprint arXiv:1806.08730, 2018.
[34]
J. Hu, Y. Yang, C. Chen, L. He, and Z. Yu, SAS: Dialogue state tracking via slot attention and slot information sharing, in Proc. 58th Annu. Meeting of the Association for Computational Linguistics, .
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 07 July 2021
Accepted: 19 July 2021
Published: 27 December 2021
Issue date: March 2022

Copyright

© The author(s) 2022

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 61976247).

Rights and permissions

The articles published in this open access journal are distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/).

Reprints and Permission requests may be sought directly from editorial office.

Return