Abstract
This dissertation presents a logic-based theory of arguments, cases, and their hardness. In this theory, we compare cases in terms of general logical formulas and define argument relations in terms of their validity strengths. We utilize the theory to study the internal structure of case models. A case model is made up of a set of cases and their preference. The theory also contributes to the study of how hard it is to decide about an issue on topics in Artificial Intelligence & Law, such as legal and evidential reasoning. The issues in our theory are conceptualized as pairs of arguments that share the same premises but opposing conclusions. For instance, we should decide whether or not to apply a legal rule in a current case following decisions in precedents. We apply the theory to model legal rules and cases as well as crime investigation in evidential reasoning. The theory demonstrates the feasibility of employing formal argumentation theory to perform AI tasks that involve combining data and knowledge, such as automated legal reasoning and legal knowledge representation. The logical foundation of the theory renders it explainable and helps mitigate human bias, thus promoting responsible AI applications in human practices and activities.
Original language | English |
---|---|
Qualification | Doctor of Philosophy |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 12-Feb-2024 |
Place of Publication | [Groningen] |
Publisher | |
DOIs | |
Publication status | Published - 2024 |