Exploiting More Associations Between Slots for Multi-Domain Dialog State Tracking
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 b...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tsinghua University Press
2022-03-01
|
| Series: | Big Data Mining and Analytics |
| Subjects: | |
| Online Access: | https://www.sciopen.com/article/10.26599/BDMA.2021.9020013 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832573644683870208 |
|---|---|
| author | Hui Bai Yan Yang Jie Wang |
| author_facet | Hui Bai Yan Yang Jie Wang |
| author_sort | Hui Bai |
| collection | DOAJ |
| description | 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. |
| format | Article |
| id | doaj-art-b81de68ad9514f6e860ccd03f507d01f |
| institution | Kabale University |
| issn | 2096-0654 |
| language | English |
| publishDate | 2022-03-01 |
| publisher | Tsinghua University Press |
| record_format | Article |
| series | Big Data Mining and Analytics |
| spelling | doaj-art-b81de68ad9514f6e860ccd03f507d01f2025-02-02T03:45:09ZengTsinghua University PressBig Data Mining and Analytics2096-06542022-03-0151415210.26599/BDMA.2021.9020013Exploiting More Associations Between Slots for Multi-Domain Dialog State TrackingHui Bai0Yan Yang1Jie Wang2School of Computing and Artifical Intelligence, Southwest Jiaotong University, Chengdu 611756, ChinaSchool of Computing and Artifical Intelligence, Southwest Jiaotong University, Chengdu 611756, ChinaSchool of Computing and Artifical Intelligence, Southwest Jiaotong University, Chengdu 611756, ChinaDialog 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.https://www.sciopen.com/article/10.26599/BDMA.2021.9020013slot-relevant attentionmulti-domain dialog state trackingtask-oriented dialog system |
| spellingShingle | Hui Bai Yan Yang Jie Wang Exploiting More Associations Between Slots for Multi-Domain Dialog State Tracking Big Data Mining and Analytics slot-relevant attention multi-domain dialog state tracking task-oriented dialog system |
| title | Exploiting More Associations Between Slots for Multi-Domain Dialog State Tracking |
| title_full | Exploiting More Associations Between Slots for Multi-Domain Dialog State Tracking |
| title_fullStr | Exploiting More Associations Between Slots for Multi-Domain Dialog State Tracking |
| title_full_unstemmed | Exploiting More Associations Between Slots for Multi-Domain Dialog State Tracking |
| title_short | Exploiting More Associations Between Slots for Multi-Domain Dialog State Tracking |
| title_sort | exploiting more associations between slots for multi domain dialog state tracking |
| topic | slot-relevant attention multi-domain dialog state tracking task-oriented dialog system |
| url | https://www.sciopen.com/article/10.26599/BDMA.2021.9020013 |
| work_keys_str_mv | AT huibai exploitingmoreassociationsbetweenslotsformultidomaindialogstatetracking AT yanyang exploitingmoreassociationsbetweenslotsformultidomaindialogstatetracking AT jiewang exploitingmoreassociationsbetweenslotsformultidomaindialogstatetracking |