On the exploration and exploitation of structural similarities in argumentative discourses

Citation: George Gkotsis, Nikos Karacapilidis (2010) On the exploration and exploitation of structural similarities in argumentative discourses. WEBIST 2010 (RSS)
Internet Archive Scholar (search for fulltext): On the exploration and exploitation of structural similarities in argumentative discourses
Tagged: Computer Science (RSS) argumentation (RSS), argument sequences (RSS), argument mining (RSS), graph theory (RSS), graphs (RSS), structure (RSS)

Summary

This paper models argumentative dialogues as a graph, in order to provide a "generic, flexible and accurate" representation of argumentation, ultimately aimed at improving argumentation support systems. The authors particular want to focus on the intermediate and late phases of argumentation (these are consensus-building; and integration-oriented and conflicted oriented consensus building, according to Weinberger and Fischer, 2006).

The key object of the paper is to apply Hay et al., 2008's vertex refinement query to arguments; understanding that work is probably key to understanding this paper. A better definition of the similarity used (i.e. whether it is intrinsic to the arguments themselves, or just about the reply structure) would make the paper more concrete and thus more understandable.

They suggest a number of (abstract) applications:

1. finding specific argumentation sequences, according to their attributes.

For instance, assuming a community rating (0 to 1) for each argument, defeated, well-supported, undersupported or ill-supported arguments could be found.

1. extracting unnoticed sequences - by searching for structural similarity

They suggest intriguing applications, such as ”show me sequences where users A, B, C attempt and fail to defeat user D's contributions”. Yet these appear to rely on either community ratings or detailed and automatic analysis of the arguments, which are not presented as a contribution of this paper.

Selected References

• Weinberger, A. and Fischer, F. (2006). A framework to analyze argumentative knowledge construction in computer-supported collaborative learning. Computers & Education, 46(1):719- 5.
• Hay, M., Miklau, G., Jensen, D., Towsley, D., and Weis, P. (2008). Resisting structural re-identification in anonymized social networks. Proceedings of the VLDB Endowment archive, 1(1):1021- 14.

Theoretical and Practical Relevance

The application of this work to real discussions relies on abstraction of particular attributes; more detail is needed (at least for this reader).