Doctoral Dissertations


"Argumentation is an important process in a collaborative decision making environment. Argumentation from a large number of stakeholders often produces a large argumentation tree. It is challenging to comprehend such an argumentation tree without intelligent analysis tools. Also, limited decision support is provided for its analysis by the existing argumentation systems. In an argumentation process, stakeholders tend to polarize on their opinions, and form polarization groups. Each group is usually led by a group leader. Polarization groups often overlap and a stakeholder is a member of multiple polarization groups. Identifying polarization groups and quantifying a stakeholder's degree of membership in multiple polarization groups helps the decision maker understand both the social dynamics and the post-decision effects on each group.

Frameworks are developed in this dissertation to identify both polarization groups and quantify a stakeholder's degree of membership in multiple polarization groups. These tasks are performed by quantifying opinions of stakeholders using argumentation reduction fuzzy inference system and further clustering opinions based on K-means and Fuzzy c-means algorithms.

Assessing the collective opinion of the group on individual arguments is also important. This helps stakeholders understand individual arguments from the collective perspective of the group. A framework is developed to derive the collective assessment score of individual arguments in a tree using the argumentation reduction inference system. Further, these arguments are clustered using argument strength and collective assessment score to identify clusters of arguments with collective support and collective attack.

Identifying outlier opinions in an argumentation tree helps in understanding opinions that are further away from the mean group opinion in the opinion space. Outlier opinions may exist from two perspectives in argumentation: individual viewpoint and collective viewpoint of the group. A framework is developed in this dissertation to address this challenge from both perspectives.

Evaluation of the methods is also presented and it shows that the proposed methods are effective in identifying polarization groups and outlier opinions. The information produced by these methods help decision makers and stakeholders in making more informed decisions"--Abstract, pages iii-iv.


Liu, Xiaoqing Frank

Committee Member(s)

Jiang, Wei
Cheng, Maggie Xiaoyan
Chellappan, Sriram
Wunsch, Donald C.


Computer Science

Degree Name

Ph. D. in Computer Science


Missouri University of Science and Technology. Intelligent Systems Center
Missouri University of Science and Technology. University Transportation Center

Research Center/Lab(s)

Intelligent Systems Center


Missouri University of Science and Technology

Publication Date



xiii, 165 pages

Note about bibliography

Includes bibliographical references (pages 151-164).


© 2013 Ravi Santosh Arvapally, All rights reserved.

Document Type

Dissertation - Open Access

File Type




Subject Headings

Decision support systems
Decision making -- Data processing
Reasoning -- Computer simulation
Logic -- Computer simulation

Thesis Number

T 10937

Print OCLC #


Electronic OCLC #