面向中文口语理解的基于依赖引导的字特征槽填充模型(A Dependency-Guided Character-Based Slot Filling Model for Chinese Spoken Language Understanding)
Zhanbiao Zhu (朱展标), Peijie Huang (黄沛杰), Yexing Zhang (张业兴), Shudong Liu (刘树东), Hualin Zhang (张华林), Junyao Huang (黄均曜)
Abstract
意图识别和槽信息填充的联合模型将口语理解技术(Spoken Language Understanding)提升到了一个新的水平,但由于存在出现频率低或未见过的槽指称项(0 shot slot mentions),模型的序列标注性能受限,而且这些联合模型往往没有利用输入序列存在的语法知识信息。已有研究表明序列标注任务可以通过引入依赖树结构,辅助推断序列标注中槽的存在。在中文口语对话理解中,由于中文话语是一串字序列组成,输入话语的字和槽信息是一一对应的,因而槽信息填充模型往往是字特征模型。基于词的依赖树结构无法直接应用于基于字特征的槽填充模型。为了解决字词之间的矛盾,本文提出了一种基于字模型的依赖引导槽填充模型(dependency guided character-based slot filling model,DCSF),提供了一种简洁的方法解决将词级依赖树结构引入中文字特征模型的冲突,同时通过对话语中词汇内部关系进行建模,保留了词级上下文信息和分词信息。在公共基准语料库当SMP-ECDT和CrossWOZ上的实验结果表明,我们的模型优于比较模型,特别是在未见过的槽指称项和低资源情况下有很大的改进。- Anthology ID:
- 2021.ccl-1.29
- Volume:
- Proceedings of the 20th Chinese National Conference on Computational Linguistics
- Month:
- August
- Year:
- 2021
- Address:
- Huhhot, China
- Editors:
- Sheng Li (李生), Maosong Sun (孙茂松), Yang Liu (刘洋), Hua Wu (吴华), Kang Liu (刘康), Wanxiang Che (车万翔), Shizhu He (何世柱), Gaoqi Rao (饶高琦)
- Venue:
- CCL
- SIG:
- Publisher:
- Chinese Information Processing Society of China
- Note:
- Pages:
- 304–315
- Language:
- Chinese
- URL:
- https://aclanthology.org/2021.ccl-1.29
- DOI:
- Cite (ACL):
- Zhanbiao Zhu, Peijie Huang, Yexing Zhang, Shudong Liu, Hualin Zhang, and Junyao Huang. 2021. 面向中文口语理解的基于依赖引导的字特征槽填充模型(A Dependency-Guided Character-Based Slot Filling Model for Chinese Spoken Language Understanding). In Proceedings of the 20th Chinese National Conference on Computational Linguistics, pages 304–315, Huhhot, China. Chinese Information Processing Society of China.
- Cite (Informal):
- 面向中文口语理解的基于依赖引导的字特征槽填充模型(A Dependency-Guided Character-Based Slot Filling Model for Chinese Spoken Language Understanding) (Zhu et al., CCL 2021)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-1/2021.ccl-1.29.pdf