Syntopical graphs for computational argumentation tasks

authored by: Joe Barrow, Rajiv Jain, Nedim Lipka, Franck Dernoncourt, Vlad I. Morariu, Varun Manjunatha, Douglas W. Oard, Philip Resnik, Henning Wachsmuth
Abstract: Approaches to computational argumentation tasks such as stance detection and aspect detection have largely focused on the text of individual claims, losing out on potentially valuable context from the broader collection of text. We present a general approach to these tasks motivated by syntopical reading, a reading process that emphasizes comparing and contrasting viewpoints in order to improve topic understanding. To capture collection-level context, we introduce the syntopical graph, a data structure for linking claims within a collection. A syntopical graph is a typed multi-graph where nodes represent claims and edges represent different possible pairwise relationships, such as entailment, paraphrase, or support. Experiments applying syntopical graphs to stance detection and aspect detection demonstrate state-of-the-art performance in each domain, significantly outperforming approaches that do not utilize collection-level information.
External Organisation(s): University of Maryland
Adobe Research
Paderborn University
Type: Conference contribution
Pages: 1583-1595
No. of pages: 13
Publication date: 2021
Publication status: Published
Peer reviewed: Yes
ASJC Scopus subject areas: Software, Computational Theory and Mathematics, Linguistics and Language, Language and Linguistics
Electronic version(s): https://doi.org/10.18653/v1/2021.acl-long.126 (Access: Open)

BibTeX

@inproceedings{bb1de3322be54b9ebeecb3ac7574f737,
title = "Syntopical graphs for computational argumentation tasks",
abstract = "Approaches to computational argumentation tasks such as stance detection and aspect detection have largely focused on the text of individual claims, losing out on potentially valuable context from the broader collection of text. We present a general approach to these tasks motivated by syntopical reading, a reading process that emphasizes comparing and contrasting viewpoints in order to improve topic understanding. To capture collection-level context, we introduce the syntopical graph, a data structure for linking claims within a collection. A syntopical graph is a typed multi-graph where nodes represent claims and edges represent different possible pairwise relationships, such as entailment, paraphrase, or support. Experiments applying syntopical graphs to stance detection and aspect detection demonstrate state-of-the-art performance in each domain, significantly outperforming approaches that do not utilize collection-level information.",
author = "Joe Barrow and Rajiv Jain and Nedim Lipka and Franck Dernoncourt and Morariu, {Vlad I.} and Varun Manjunatha and Oard, {Douglas W.} and Philip Resnik and Henning Wachsmuth",
note = "Funding Information: We would like to thank many others for their invaluable feedback and patient discussions, including Charlotte Ellison, Ani Nenkova, Tong Sun, Han-Chin Shing, and Pedro Rodriguez. This work was generously supported through Adobe Gift Funding, which supports an Adobe Research-University of Maryland collaboration. It was completed while the primary author was interning at Adobe Research.; Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, ACL-IJCNLP 2021 ; Conference date: 01-08-2021 Through 06-08-2021",
year = "2021",
doi = "10.18653/v1/2021.acl-long.126",
language = "English",
isbn = "9781954085527",
pages = "1583--1595",
booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing",
publisher = "Association for Computational Linguistics (ACL)",
}

Publication Details

Syntopical graphs for computational argumentation tasks

Funded by