Abstract
Dialog state tracking (DST) is the key component of goal-driven Spoken Dialog Systems. Almost all existing dialog state trackers are unable to handle unknown slot values. The continuous emergence of unknown slot values in dialogs is inevitable. If unknown slot values cannot be accurately detected and distinguished, the dialog states cannot be correctly updated in real time. This paper proposes a hierarchical dialog state tracking framework to model the dialog state tracking with unknown slot values. Three levels are included in the framework. Unknown slot values are identified at the first-level by a two-layer cascaded neural network as well as known slot values. Distributions for unknown and known slot values are updated separately in the second level and integrated in the third level. Experimental results on DSTC and WOZ2.0 datasets show that the proposed framework achieves good performance. Especially, the detection and distinction of unknown slot values greatly improve the final performance of dialog state tracking, illustrating the effectiveness of our proposed framework for addressing the problem of unknown slot values.
Similar content being viewed by others
References
Williams J, Raux A, Henderson M (2016) The dialog state tracking challenge series: a review. Dialogue Discourse 7(3):4–33
Henderson M, Thomson B, Williams J (2014) The second dialog state tracking challenge. In: SIGDIAL, pp 263–272
Henderson M, Thomson B, Williams JD (2014) The third dialog state tracking challenge. In: SLT workshop, pp 324–329
Williams JD, Poupart P, Young S (2005) Factored partially observable markov decision processes for dialogue management. In: Proceedings of the workshop on knowledge and reasoning in practical dialog systems, IJCAI, pp 76-82
Young S, Gašić M, Thomson B, Williams JD (2013) Pomdp-based statistical spoken dialog systems: a review. Proc IEEE 101(5):1160–1179
Bohus D, Rudnicky A (2006) A ’k hypotheses + other belief updating model. In: Proceedings of the AAAI workshop on statistical and empirical approaches for spoken dialogue systems
Metallinou A, Bohus D, Williams J (2013) Discriminative state tracking for spoken dialog systems. In: ACL, pp 466–475
Williams JD (2014) Web-style ranking and slu combination for dialog state tracking. In: SIGDIAL, pp 282–291
Ren H, Xu WQ, Yan YH (2014) Markovian discriminative modeling for cross-domain dialog state tracking. In: SLT workshop, pp 342–347
Kim S, Banchs RE (2014) Sequential labeling for tracking dynamic dialog states. In: SIGDIAL, pp 332–336
Henderson M, Thomson B, Young S (2014) Word-based dialog state tracking with recurrent neural networks. In: SIGDIAL, pp 292–299
Mrkšić N, Séaghdha DÓ, Wen TH, Thomson B, Young S (2017) Neural belief tracker: data-driven dialogue state tracking. In: ACL, pp 1777–1788
Yang XH, Liu J (2015) Dialog state tracking using long short-term memory neural networks. INTERSPEECH 2015:1800–1804
Shi HJ, Ushio T, Endo M, Yamagami K, Horii N (2017) Convolutional neural networks for multi-topic dialog state tracking. Springer, Berlin
Jang Y, Ham J, Lee BJ, Chang Y, Kim KE (2017) Neural dialog state tracker for large ontologies by attention mechanism. In: SLT workshop, pp 531–537
Mrkšić N, Séaghdha DÓ, Thomson B, Gašić M, Su PH, Vandyke D, Wen TH, Young S (2015) Multidomain dialog state tracking using recurrent neural networks. In: ACL, pp 794–799
Lee BJ, Kim KE (2016) Dialog history construction with long-short term memory for robust generative dialog state tracking. Dialogue Discourse 7(3):47–64
Sun K, Chen L, Zhu S, Yu K (2014) The sjtu system for dialog state tracking challenge 2. In: SIGDIAL, pp 318–326
Henderson M, Thomson B, Young S (2015) Robust dialog state tracking using delexicalised recurrent neural networks and unsupervised adaptation. In: SLT workshop, pp 360–365
Yazdani M, Henderson J (2015) A model of zero-shot learning of spoken language understanding. In: EMNLP, pp 244-249
Ferreira E, Jabaian B, Lefevr F (2015) Online adaptative zero-shot learning spoken language understanding using word-embedding. In: ICASSP, pp 5321–5325
Mu X, Zhu F, Du J, Lim EP, Zhou ZH (2017) Streaming classification with emerging new class by class matrix sketching. In: AAAI
Kadlec R, Vodolán M, Libovický J, Macek J, Kleindienst J (2014) Knowledge-based dialog state tracking. In: SLT workshop, pp 348–353
Gers F, Schmidhuber J, Cummins F (2000) Learning to forget: continual prediction with lstm. Neural Comput 12(10):2451–2471
Kalchbrenner N, Grefenstette E, Blunsom P (2014) A convolutional neural network for modelling sentences. In: ACL
Kim S, Banchs RE, Li HZ (2016) Exploring convolutional and recurrent neural networks in sequential labelling for dialogue topic tracking. In: ACL, pp 963–973
Zhang S, Choromanska A, LeCun Y (2015) Deep learning with elastic averaging sgd. In: NIPS, pp 685–693
Mrkšić N, Séaghdha DÓ, Thomson B, Gašić M, Rojas-Barahona L, Su PH, Vandyke D, Wen TH, Young S (2016) Counter-fitting word vectors to linguistic constraints. In: Conference of the North American Chapter of the association for computational linguistics: human language technologies, pp 142–148
Acknowledgements
This paper is supported by NSFC (No. 61273365), NSSFC (2016ZDA055), 111Project (No. B08004), Beijing Advanced Innovation Center for Imaging Technology, Engineering Research Center of Information Networks of MOE, China.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Yang, G., Wang, X. & Yuan, C. Hierarchical Dialog State Tracking with Unknown Slot Values. Neural Process Lett 50, 1611–1625 (2019). https://doi.org/10.1007/s11063-018-9950-1
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-018-9950-1