Abstract |
Argumentation is an important human activity concerning many aspects of human lifeand society such as politics, law, scientific and scholarly discourse. Nowadays, more and more people communicate and publish arguments via the Internet. For this reason, it is of
vital importance to build information systems that manage, structure and fully understand human argumentation processes. In recent years, significant efforts have been made in
connecting Information Technology with argumentation. However, current argumentation systems are based on models that are either not analytical enough or restricted to formal
logic. So far, none of these models connects argumentation with a domain ontology. A small minority of these systems use Semantic Web technologies for analyzing structuring
and representing arguments. Furthermore, most of these models blur the logical structure of a composite argument that is believed at some point in time with the temporal order of
arguments in the argumentation process.
In this thesis we describe an integrated model for human argumentation in which reasoning may not only consist of falsification or verification but more generally of strengthening
or weakening hypotheses. The model includes evolution, composition and revision of arguments and can be connected to a domain ontology. We explain the extensions that were
made in order to specialize this model in factual argumentation, i.e., in arguments and counterarguments about propositions concerning material states of affairs in the past. Also,
we introduce the extensions that were made in the model in order to be able to defeat arguments that cause inconsistencies. Next, we describe the implementation of the model
using the Resource Description Framework Schema (RDFS) semantic language. We present Antilogos, a Semantic Web-based argumentation system for the record and representation of dynamic arguments networks. The aim of the system is to inform the users about the provenance of the registered knowledge in a structured way. Moreover, the system is able 5 to detect and resolve con°icts between the arguments. Antilogos enables users to create new arguments and also to support, defeat and search propositions of existing arguments.
Through Antilogos, users can monitor any state of the knowledge through time and can also understand how this knowledge came up or how was changed. Finally, we demonstrate
the use of Antilogos by presenting arguments and counter-arguments of a real published archaeological case study.
|