National and Kapodistrian University of Athens
Αυτόματες διαπραγματεύσεις υπολογιστικά νοημόνων οντοτήτων σε ηλεκτρονικές αγορές
Abstract
dc:descriptionAutomated negotiations consist an interesting research domain for many years. A scenario, mostly depicting real life negotiations, defines that entities act under no knowledge on the characteristics of the rest of them. This means that their behavior should incorporate mechanisms for handling uncertainty created by the lack of knowledge as well as intelligent methods for modelling every aspect of the discussed scenario. In this PhD Thesis, we adopt computational intelligence techniques in order to propose efficient mechanisms for the definition of the behavior of entities participating in Electronic Markets. We cover the entire framework defined in a marketplace by proposing methodologies for the definition of basic parameters together with decision making models at every step of each negotiation. We take into consideration the uncertainty in such scenarios together with profit maximization. We propose decision making models that are based on different aspects of the discussed scenario in order to reveal the optimal one. We study buyers’ as well as sellers’ behavior. The proposed decision making mechanisms, for every part (buyers and sellers), are based on Fuzzy Logic, Game Theory, Swarm Intelligence and Optimal Stopping Theory.We describe methods for the selection of middle entities and products. The proposed methods result the appropriate middle entity or product that best matches the buyers’ needs. In these methods, we utilize Quality of Service parameters in order to increase the efficiency of the proposed model. We study negotiations between one buyer and one seller as well as concurrent negotiations between a buyer and multiple sellers. In the first case, we are based on the Game Theory principles with the final aim to provide a model that maximizes the expected profit. For the second case, we are based on Swarm Intelligence theory in order to have a framework where threads, used by the buyer, conclude the best solution (the best agreement) through a team work. Additionally, Optimal Stopping Theory gives us the tools to change the view of the problem. Based on Optimal Stopping Theory, we propose models trying to find the best time to take a decision.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2013
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kolomvatsos, Konstantinos
- Κολομβάτσος, Κωνσταντίνος
Subjects
dc:subject × 20- Αυτόματες διαπραγματεύσεις
- Αυτόνομες οντότητες
- Ασαφής λογική
- Θεωρία του σμήνους
- Θεωρία βέλτιστης παύσης
- Θεωρία παιγνίων
- Νευρωνικά δίκτυα
- Ενισχυτική μάθηση
- Automated negotiations
- Autonomous entities
- Fuzzy logic
- Swarm intelligence
- Optimal stopping theory
- Game theory
- Neural networks
- Reinforcement learning
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Natural Sciences
- Computer and Information Sciences
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/38120