ArguNet: Exploiting Dialogue Acts and Context to Identify Argumentative Relations in Online Debates

Anonymous

TL;DR: This paper introduces a neural model for argumentative relation classification in online debate data.

Abstract: Argumentative Relation Classification is the task of determining the relationship between two arguments within the context of an argumentative dialogue. Existing models in the literature rely on a combination of lexical features and pre-trained Large Language Models (LLMs) to tackle this task; while this approach is somewhat effective, it fails to take into account the importance of pragmatic features such as the illocutionary force of the argument or the structure of previous utterances in the discussion. In this work, we introduce ArguNet, a new model for Argumentative Relations Classification which relies on a combination of Dialogue Acts and Dialogue History to obtain a more nuanced understanding of an argument's stance. We show that our model achieves state-of-the-art results on the Kialo benchmark test set, and provide evidence of its robustness in an open-domain scenario.

Paper Type: long

Research Area: Sentiment Analysis, Stylistic Analysis, and Argument Mining

Contribution Types: NLP engineering experiment, Reproduction study

Languages Studied: English