Analyzing Credibility of Arguments in a Web-Based Intelligent Argumentation System for Collective Decision Support based on K-means Clustering Algorithm

Abstract

We developed an intelligent argumentation and collaborative decision support system which allows stakeholders to exchange arguments and captures their rationale. Arguments with lack of credibility in an argumentation tree may negatively affect decisions in a collaborative decision making process if they are not identified collectively by the group. to address this issue, we perform clustering analysis on an argumentation tree using K-means clustering algorithm on credibility factors of arguments such as degree of an argument, and collective determination of an argument. Arguments are classified into multiple groups: from highly credible to lack of credibility. It helps capture rationale of selection of the most favorable solution alternative by the system. It helps decision makers identify arguments with high credibility based on collective determination. We perform an empirical study of the method and its results indicate that it is effective in supporting collective decision making using the system. © 2012 Operational Research Society. All rights reserved.

Department(s)

Computer Science

Keywords and Phrases

Argumentation systems; Collaborative systems; Collective intelligence; Decision support; Group decision support; Machine learning algorithms

International Standard Serial Number (ISSN)

1477-8246; 1477-8238

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2024 Taylor and Francis Group; Taylor and Francis; OR Society, All rights reserved.

Publication Date

01 Jan 2012

Link to Full Text

Share

 
COinS