1. Αποθετήριο Παραδοτέων Υποτρόφων Δράσεων ΕΣΠΑ Τριτοβάθμιας Εκπαίδευσης 2017-2023
2. Όλο το Αποθετήριο
3. Ερευνητικές ομάδες

Τύπος:	Κείμενο εργασίας
Τίτλος:	Applications of computational techniques on aspects of Social Choice Theory
Συγγραφέας:	[EL] Καρανικόλας, Νικόλαος[EN] Karanikolas, Nikos
Επιβλέπων διατριβής:	[EL] Κακλαμάνης, Χρήστος[EN] Kaklamanis, Christos
Ημερομηνία:	Απρ-2019
Περίληψη:	During this post-doctoral research we focus on applying computational techniques for solving problems and issues of Social Choice Theory. More precisely, we deepen on problems arising in collective decision-making under a social choice theoretic approach. That is, we adopt methods and techniques that lie on aspects of social choice theory for aggregating preferences with aim to aid collective decision making. In addition, we investigate the way of enriching the classical social choice approach by examining the effect of reasoning in the preferences of the group when a collective decision is to be made. One main aspect of social choice theory is preference aggregation, which in general deals with the aggregation of individual agents’ preferences into a collective preference. In a more specific context, a prominent topic of social choice is voting theory, which constitutes a form of preference aggregation. The central problem of voting theory is the computation of the winning alternative in an election when we have as input the preferences of the voters, i.e., the agents. In the literature many voting rules have been proposed in order to aggregate the agents’ preferences having as an objective the computation of the winner of the elections. Voting theory is a seminal subject in the computational social choice theory with applications in the society as we can see that voting rules are widely used in collective decision making. For example, voting rules are used in government and municipal elections, groups and committees for taking decisions, voting polls across the internet, etc. During our research we focus on aspects of collective decision-making when social choice methods are applied on a group of decision makers, which we call them as agents. We start by considering collective decision making and preference aggregation mechanisms, which take as input the preferences of the agents as well as the provided reasoning for these preferences. Our research hypothesis is that a decision made by a group of agents understanding the qualitative rationale behind each other’s preferences has better chances to be accepted and used in practice. We follow an algorithmic approach and propose two collective decision making methods which combine argumentation and computational social choice techniques. The first one is a novel qualitative decision process while the other method is a quantitative one. For the qualitative approach we prove theoretical results which show that it can overcome some of the social choice deficiencies. In the quantitative method we quantify the deliberation phase by defining a new voting argumentation framework and its acceptability semantics. We prove theoretical results for these semantics regarding well-known properties that appear in argumentation and social choice theory. Next, we focus on the problem of multi-criteria decision making, where the goal is to reach an acceptable collective decision aggregating agents’ preferences expressed over multiple criteria. We provide a novel modelling of the multi-criteria decision making problem as an inconsistent knowledge base, and we explain how to benefit from the reasoning capabilities of existential rules. The repairs of this knowledge base represent the maximally consistent point of views and inference strategies can be used for decision making. In the last phase of our research we turn our attention in real decision making applications by implementing aggregation algorithms following a visualizing approach. First, we design a decision-making software tool that applies in decision problems of agricultural engineering. The tool uses methods of computational social choice and argumentation for preference aggregation and collective decision making. Hence, the tool’s architecture is composed of two main systems, i.e., the social choice system and the deliberation system. We mainly focus on the social choice system where its implementation is oriented towards practicality, so that it can be applied to different decision problems used in agriculture and are related with the valorization of materials. We complete the picture by providing the implementation of all the algorithms used and needed to support the theoretical results of this research. These algorithms consist the core of future work regarding software tools that use reasoning based methods for collective decision making.
Γλώσσα:	Αγγλικά
Τόπος δημοσίευσης:	Πάτρα, Ελλάδα
Σελίδες:	113
Θεματική κατηγορία:	[EL] Επιστήμη ηλεκτρονικών υπολογιστών[EN] Computer Science
Κάτοχος πνευματικών δικαιωμάτων:	© The Author(s) 2019
Σημειώσεις:	The current research was implemented with a scholarship from IKY funded by the action “Support of Postdoctoral Researchers” from the resources of the EP “Human Resources Development, Education and Lifelong Learning” with priority axes 6, 8, 9 and is co-funded by the European Social Fund - ESF and the Greek state.
Εμφανίζεται στις συλλογές:	Ερευνητικές ομάδες