Identifying the Human Values behind Arguments

39075

Authors

nan

Abstract

AbstractThis paper studies the (often implicit) human values behind natural language arguments, such as to have freedom of thought or to be broadminded. Values are commonly accepted answers to why some option is desirable in the ethical sense and are thus essential both in real-world argumentation and theoretical argumentation frameworks. However, their large variety has been a major obstacle to modeling them in argument mining. To overcome this obstacle, we contribute an operationalization of human values, namely a multi-level taxonomy with 54 values that is in line with psychological research. Moreover, we provide a dataset of 5270 arguments from four geographical cultures, manually annotated for human values. First experiments with the automatic classification of human values are promising, with F1-scores up to 0.81 and 0.25 on average.

Keywords

nan

Full Content

Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics Volume 1: Long Papers, pages 4459 - 4471 May 22-27, 2022 c⃝2022 Association for Computational Linguistics Identifying the Human Values behind Arguments Johannes Kiesel Bauhaus-Universität Weimar johannes.kiesel@uni-weimar.de Milad Alshomary Paderborn University milad.alshomary@upb.de Nicolas Handke Universität Leipzig nicolas.handke@gmx.de Xiaoni Cai Technische Universität München caix@in.tum.de Henning Wachsmuth Paderborn University henningw@upb.de Benno Stein Bauhaus-Universität Weimar benno.stein@uni-weimar.de Abstract This paper studies the (often implicit) human values behind natural language arguments, such as to have freedom of thought or to be broad- minded. Values are commonly accepted an- swers to why some option is desirable in the ethical sense and are thus essential both in real- world argumentation and theoretical argumen- tation frameworks. However, their large va- riety has been a major obstacle to modeling them in argument mining. To overcome this obstacle, we contribute an operationalization of human values, namely a multi-level taxonomy with 54 values that is in line with psychologi- cal research. Moreover, we provide a dataset of 5270 arguments from four geographical cul- tures, manually annotated for human values. First experiments with the automatic classifi- cation of human values are promising, with F1-scores up to 0.81 and 0.25 on average. 1 Introduction How come people disagree on the best course for- ward in controversial issues, even if they use the same information to form their opinion? A way to get to the bottom of such disagreement is to repeatedly ask them why they see something as desirable. We observe that people have different beliefs and priorities of what is generally worth striving for (e.g., personal achievements vs. humil- ity) and how to do so (e.g., being self-directed vs. respecting traditions), often referred to as (human) values (Searle, 2003). Some values tend to conflict and others to align (see Figure 1), which can cause disagreement on the best course forward, but also the support, if not formation, of political parties that promote the respective highly revered values. Moreover, one can observe different value priori- ties between cultures and disagreement thereon. Level 1: Values C o n s e r v at io n Humility Hedonism Face G r o w t h , A n xi et y- fr e e P e r s o n a l f o c u s S o c i a l f o c u s O p e n n e s s t o c h a n g e S el f- pr o te ct io n, A n xi et y- a v oi d a n c e S el f- di re ct io n: C o nf or m it y: Stimulation thought action Achievement dominance resources personal societal Tradition rules interpersonal caring concern nature tolerance objectivity dependability B e n e v ol e n c e: U ni v er s al is m : S el f- tr a n s c e n d e n c e S …

Content truncated to 500 words

Labeled Text

AbstractThis paper studies the (often implicit) human values behind natural language arguments, such as to have freedom of thought or to be broadminded. Values are commonly accepted answers to why some option is desirable in the ethical sense and are thus essential both in real-world argumentation and theoretical argumentation frameworks. However, their large variety has been a major obstacle to modeling them in argument mining. To overcome this obstacle, we contribute an operationalization of human values, namely a multi-level taxonomy with 54 values that is in line with psychological research. Moreover, we provide a dataset of 5270 arguments from four geographical cultures, manually annotated for human values. First experiments with the automatic classification of human values are promising, with F1-scores up to 0.81 and 0.25 on average.

Publication Details

Paper ID:
39075
Year:
2022
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan