TU Darmstadt / ULB / TUbiblio

Serious Games for large-scale Argumentation Mining

Hannemann, Raffael (2015):
Serious Games for large-scale Argumentation Mining.
Technische Universität Darmstadt, [Online-Edition: https://download.hrz.tu-darmstadt.de/media/FB20/Dekanat/Publ...],
[Master Thesis]

Abstract

Communicating a point of view and convincing a conversation partner with a clear, sophisticated line of argument is a generally demanded skill nowadays. Also, balancing pros and cons plays an important role in the daily decision making process. Thus, generally a high importance can be attributed to the theory of argumentation. There is a need for structured, annotated corpora emerging from the still young domain of computer based analysis of argumentative texts and the computer assisted writing. New approaches in the domain of Machine Learning, Artificial Intelligences and Data Mining demand such types of data sets. The characteristics of the required corpus data may demand very distinct requirements, which often implies individually tailored solutions for the use cases. However, since the manual creation of annotated corpora consumes a lot of time and labour resources, the research for novel, alternative approaches is highly appreciated. This thesis proposes a novel approach at the interface of Argumentation Mining and Serious Games to collect a large corpus of structured, high quality arguments by leveraging a game based approach with crowd-sourcing elements. The developed game is an application for browsers and smartphones which motivates the users to both generate and analyze arguments for current hot topics. The proposed game mechanics cultivate the active participation in the creation process, and furthermore have integrated a mutual validation of the generated data by the users themselves to ensure the quality of the resulting corpus. In the sense of didactic games, the proposed game furthermore imparts the foundations of argumentation theory, which provides additional incentive to participate. Finally, to examine particular aspects of the developed concept, a user study has been conducted and its results have been evaluated.

Item Type: Master Thesis
Erschienen: 2015
Creators: Hannemann, Raffael
Title: Serious Games for large-scale Argumentation Mining
Language: English
Abstract:

Communicating a point of view and convincing a conversation partner with a clear, sophisticated line of argument is a generally demanded skill nowadays. Also, balancing pros and cons plays an important role in the daily decision making process. Thus, generally a high importance can be attributed to the theory of argumentation. There is a need for structured, annotated corpora emerging from the still young domain of computer based analysis of argumentative texts and the computer assisted writing. New approaches in the domain of Machine Learning, Artificial Intelligences and Data Mining demand such types of data sets. The characteristics of the required corpus data may demand very distinct requirements, which often implies individually tailored solutions for the use cases. However, since the manual creation of annotated corpora consumes a lot of time and labour resources, the research for novel, alternative approaches is highly appreciated. This thesis proposes a novel approach at the interface of Argumentation Mining and Serious Games to collect a large corpus of structured, high quality arguments by leveraging a game based approach with crowd-sourcing elements. The developed game is an application for browsers and smartphones which motivates the users to both generate and analyze arguments for current hot topics. The proposed game mechanics cultivate the active participation in the creation process, and furthermore have integrated a mutual validation of the generated data by the users themselves to ensure the quality of the resulting corpus. In the sense of didactic games, the proposed game furthermore imparts the foundations of argumentation theory, which provides additional incentive to participate. Finally, to examine particular aspects of the developed concept, a user study has been conducted and its results have been evaluated.

Uncontrolled Keywords: UKP_a_ArMin
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Ubiquitous Knowledge Processing
Date Deposited: 31 Dec 2016 14:29
Official URL: https://download.hrz.tu-darmstadt.de/media/FB20/Dekanat/Publ...
Identification Number: TUD-CS-2015-0108
Referees: Habernal, Ivan and Stab, Christian
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item