Reports of personal experiences and stories in argumentation: datasets and analysis

Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics Volume 1: Long Papers, pages 5530 - 5553 May 22-27, 2022 c⃝2022 Association for Computational Linguistics Reports of personal experiences and stories in argumentation: datasets and analysis Neele Falk and Gabriella Lapesa Institute for Natural Language Processing, University of Stuttgart {neele.falk,gabriella.lapesa}@ims.uni-stuttgart.de Abstract Reports of personal experiences or stories can play a crucial role in argumentation, as they represent an immediate and (often) relatable way to back up one’s position with respect to a given topic. They are easy to understand and increase empathy: this makes them powerful in argumentation. The impact of personal re- ports and stories in argumentation has been studied in the Social Sciences, but it is still largely underexplored in NLP. Our work is the first step towards filling this gap: our goal is to develop robust classifiers to identify docu- ments containing personal experiences and re- ports. The main challenge is the scarcity of an- notated data: our solution is to leverage exist- ing annotations to be able to scale-up the analy- sis. Our contribution is two-fold. First, we con- duct a set of in-domain and cross-domain ex- periments involving three datasets (two from Argument Mining, one from the Social Sci- ences), modeling architectures, training setups and fine-tuning options tailored to the involved domains. We show that despite the differences among datasets and annotations, robust cross- domain classification is possible. Second, we employ linear regression for performance min- ing, identifying performance trends both for overall classification performance and individ- ual classifier predictions. 1 Introduction Although personal narratives and experiences natu- rally fill an important place in our everyday discus- sions, they still do not conform to the classic ideal of a “good” argument. A “good” argument con- tains facts and logical conclusions, but as soon as more personal or emotional nuances are involved, it deviates from the norm. According to the the- ory of Deliberative Democracy (Habermas, 1996; Fishkin, 1995), the discourse that precedes political decisions plays a central role. Here, too, until the so-called affective turn in Social Sciences (Hoggett and Thompson, 2012), the assumption was that an exchange of arguments that is as rational as possi- ble can lead to better political decisions. Recent studies in Deliberative Theory, however, are in- creasingly concerned with the interplay between classical argumentation and alternative forms of ar- gumentation, which include personal experiences, narratives and emotions, and their positive effects on discourse (Polletta and Lee, 2006; Esau, 2018; Gerber et al., 2018; Maia et al., 2020). A norm that only allows rational and logical argumenta- tion firstly does not correspond to human realis- tic communication and secondly bears the danger that less educated groups or groups in which other communication standards prevail are marginalized. Arguments with personal experiences are therefore important to fulfill one of the core deliberative stan- dards, namely inclusivity (Polletta and Gardner, 2018). The following example is taken from a discus- sion about regulations to ban peanut products on airlines and illustrates further …

Content truncated to 500 words