Using argumentation to improve classification in natural language problems
Author(s)
Carstens, L
Toni, F
Type
Journal Article
Abstract
Argumentation has proven successful in a number of domains, including Multi-Agent Systems and decision support
in medicine and engineering. We propose its application to a domain yet largely unexplored by argumentation re-
search: Computational linguistics. We have developed a novel classification methodology that incorporates reasoning
through argumentation with supervised learning. We train classifiers and then
argue
about the validity of their out-
put. To do so we identify arguments that formalise prototypical knowledge of a problem and use them to correct
misclassifications. We illustrate our methodology on two tasks. On the one hand we address
cross-domain sentiment
polarity classification
, where we train classifiers on one corpus, e.g. Tweets, to identify positive/negative polarity,
and classify instances from another corpus, e.g. sentences from movie reviews. On the other hand we address a form
of argumentation mining that we call
Relation-based Argumentation Mining
, where we classify pairs of sentences
based on whether the first sentence attacks or supports the second, or whether it does neither. Whenever we find
that one sentence attacks/supports the other we consider both to be argumentative, irrespective of their stand-alone
argumentativeness. For both tasks we improve classification performance when using our methodology, compared to
using standard classifiers only.
in medicine and engineering. We propose its application to a domain yet largely unexplored by argumentation re-
search: Computational linguistics. We have developed a novel classification methodology that incorporates reasoning
through argumentation with supervised learning. We train classifiers and then
argue
about the validity of their out-
put. To do so we identify arguments that formalise prototypical knowledge of a problem and use them to correct
misclassifications. We illustrate our methodology on two tasks. On the one hand we address
cross-domain sentiment
polarity classification
, where we train classifiers on one corpus, e.g. Tweets, to identify positive/negative polarity,
and classify instances from another corpus, e.g. sentences from movie reviews. On the other hand we address a form
of argumentation mining that we call
Relation-based Argumentation Mining
, where we classify pairs of sentences
based on whether the first sentence attacks or supports the second, or whether it does neither. Whenever we find
that one sentence attacks/supports the other we consider both to be argumentative, irrespective of their stand-alone
argumentativeness. For both tasks we improve classification performance when using our methodology, compared to
using standard classifiers only.
Date Issued
2017-07-14
Date Acceptance
2016-11-14
Citation
ACM Transactions on Internet Technology, 2017, 17 (3)
ISSN
1557-6051
Publisher
Association for Computing Machinery (ACM)
Journal / Book Title
ACM Transactions on Internet Technology
Volume
17
Issue
3
Copyright Statement
© 2017 ACM. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in ACM Transactions on Internet Technology (TOIT), (July 2017) http://dl.acm.org/citation.cfm?doid=3106680.3017679
Subjects
Networking & Telecommunications
0805 Distributed Computing
0806 Information Systems
0801 Artificial Intelligence And Image Processing
Publication Status
Published
Article Number
30