Towards Argument Mining for Social Good: A Survey

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, pages 1338–1352 August 1–6, 2021. ©2021 Association for Computational Linguistics 1338 Towards Argument Mining for Social Good: A Survey Eva Maria Vecchi, Neele Falk, Iman Jundi and Gabriella Lapesa Institute for Natural Language Processing University of Stuttgart (Germany) first[-middle].last@ims.uni-stuttgart.de Abstract This survey builds an interdisciplinary picture of Argument Mining (AM), with a strong fo- cus on its potential to address issues related to Social and Political Science. More specifically, we focus on AM challenges related to its ap- plications to social media and in the multilin- gual domain, and then proceed to the widely debated notion of argument quality. We pro- pose a novel definition of argument quality which is integrated with that of deliberative quality from the Social Science literature. Un- der our definition, the quality of a contribu- tion needs to be assessed at multiple levels: the contribution itself, its preceding context, and the consequential effect on the develop- ment of the upcoming discourse. The latter has not received the deserved attention within the community. We finally define an applica- tion of AM for Social Good: (semi-)automatic moderation, a highly integrative application which (a) represents a challenging testbed for the integrated notion of quality we advocate, (b) allows the empirical quantification of argu- ment/deliberative quality to benefit from the developments in other NLP fields (i.e. hate speech detection, fact checking, debiasing), and (c) has a clearly beneficial potential at the level of its societal thanks to its real-world ap- plication (even if extremely ambitious). 1 Introduction Considering Argument Mining (AM) for Social Good implies a strong conceptual shift: the dis- course exchange is not to be interpreted as a com- petition to be won by the most persuasive contribu- tion1, but rather as a cooperative endeavor in which 1In this paper, we use the term ”contribution” to refer to a turn in a discourse exchange; more concretely a contribution is a textual unit in a discourse contex, e.g., a post in a forum, a tweet in a discussion thread; a speech in a parliamentary debate). each individual contribution represents a move to- wards a shared goal. If argumentative discourse is cooperation, it is not to be taken for granted that the perfect debater, most often the primary objective in AM research, is necessarily also the best team player. Building on this assumption, we review recent developments in the field of AM from the perspec- tive of its application in socially relevant contexts. Our survey has a strong interdisciplinary perspec- tive, putting the focus on the collaboration between NLP and the Social Sciences and, more specifi- cally, in argumentation targeted at decision-making (deliberation). Deliberative discourse historically characterizes parliamentary debates; however, it pervades, more and more frequently, discussions in digital democracy forums and, beyond that, spe- cific strands of discussions in “generalistic” social media. Looking at argumentation through the lens of deliberation has a 2-fold benefit. From a …

Content truncated to 500 words