多轮对话的篇章级抽象语义表示标注体系研究(Research on Discourse-level Abstract Meaning Representation Annotation framework in Multi-round Dialogue)
Tong Huang (黄彤), Bin Li (李斌), Peiyi Yan (闫培艺), Tingting Ji (计婷婷), Weiguang Qu (曲维光)
Abstract
对话分析是智能客服、聊天机器人等自然语言对话应用的基础课题,而对话语料与常规书面语料有较大差异,存在大量的称谓、情感短语、省略、语序颠倒、冗余等复杂现象,对句法和语义分析器的影响较大,对话自动分析的准确率相对书面语料一直不高。其主要原因在于对多轮对话缺乏严整的形式化描写方式,不利于后续的分析计算。因此,本文在梳理国内外针对对话的标注体系和语料库的基础上,提出了基于抽象语义表示的篇章级多轮对话标注体系。具体探讨了了篇章级别的语义结构标注方法,给出了词语和概念关系的对齐方案,针对称谓语和情感短语增加了相应的语义关系和概念,调整了表示主观情感词语的论元结构,并对对话中一些特殊现象进行了规定,设计了人工标注平台,为大规模的多轮对话语料库标注与计算研究奠定基础。- Anthology ID:
- 2020.ccl-1.6
- Volume:
- Proceedings of the 19th Chinese National Conference on Computational Linguistics
- Month:
- October
- Year:
- 2020
- Address:
- Haikou, China
- Editors:
- Maosong Sun (孙茂松), Sujian Li (李素建), Yue Zhang (张岳), Yang Liu (刘洋)
- Venue:
- CCL
- SIG:
- Publisher:
- Chinese Information Processing Society of China
- Note:
- Pages:
- 54–65
- Language:
- Chinese
- URL:
- https://aclanthology.org/2020.ccl-1.6
- DOI:
- Cite (ACL):
- Tong Huang, Bin Li, Peiyi Yan, Tingting Ji, and Weiguang Qu. 2020. 多轮对话的篇章级抽象语义表示标注体系研究(Research on Discourse-level Abstract Meaning Representation Annotation framework in Multi-round Dialogue). In Proceedings of the 19th Chinese National Conference on Computational Linguistics, pages 54–65, Haikou, China. Chinese Information Processing Society of China.
- Cite (Informal):
- 多轮对话的篇章级抽象语义表示标注体系研究(Research on Discourse-level Abstract Meaning Representation Annotation framework in Multi-round Dialogue) (Huang et al., CCL 2020)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2020.ccl-1.6.pdf