arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.14647cs.CL
收藏

基于论证语义学与自然语言论证图网络的自动辩论评估

Automatic Debate Evaluation with Argumentation Semantics and Natural Language Argument Graph Networks

Centre for Argument Technology, University of Dundee(邓迪大学论证技术中心);VRAIN, Universitat Politècnica de València(瓦伦西亚理工大学VRAIN研究所)

Ramon Ruiz-Dolz, Stella Heras, Ana García-Fornes

AI总结:

针对辩论标注数据不足导致自动评估困难的问题,本文融合论证理论、Transformer 与神经图网络,提出混合式自动辩论评估方法并验证其有效性。

摘要:

专业论证和完整论证性辩论方面标注数据的缺乏,导致相关自然语言处理任务被过度简化,且难以处理更复杂的任务,自动辩论评估便是如此。本文提出一种原创的混合方法,用于自动评估论证性辩论。为此,我们将论证框架和语义学等论证理论概念,与基于 Transformer 的架构及神经图网络相结合。此外,我们获得了富有前景的结果,为自然语言论证自动分析中一个尚未探索的新实例奠定了基础。

英文摘要:

The lack of annotated data on professional argumentation and complete argumentative debates has led to the oversimplification and the inability of approaching more complex natural language processing tasks. Such is the case of the automatic debate evaluation. In this paper, we propose an original hybrid method to automatically evaluate argumentative debates. For that purpose, we combine concepts from argumentation theory such as argumentation frameworks and semantics, with Transformer-based architectures and neural graph networks. Furthermore, we obtain promising results that lay the basis on an unexplored new instance of the automatic analysis of natural language arguments.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑