National Technical University of Athens (NTUA)
Ολοκληρωμένη μεθοδολογία και πληροφοριακό σύστημα υποστήριξης ομαδικών αποφάσεων ταξινόμησης με την χρήση πολυκριτηριακής ανάλυσης: Εφαρμογή στον Τραπεζικό Τομέα
Abstract
dc:descriptionUtilization of multicriteria analysis in group decision support has been extended during the last decade, despite the inherent complexity of the field. However, the majority of methodologies focus on choice and ranking problems, while there is no methodology available for classification ones. In addition to methodological lack in theoretical level, there also exists lack in the application domain, especially regarding the design and development of appropriate Group Decision Support Systems (GDSS) for classification problems. Considering the above, as well as relevant needs in the financial domain, the present dissertation focuses on covering both the theoretical and the application lack. More specifically, its research domain lies within the field of small collaborative group decision making for multicriteria classification problems in predefined non ordered categories. In details, NeXClass multicriteria methodology is presented, which is a novel approach and supports classification problems for predefined non ordered categories by a single decision maker. The methodology introduces the definition of categories according to their limits, which is considered as entrance threshold, and the fuzzy inclusion degree, on which classification is based. NeXClass is extended in group setting and the novel methodology NeXClass-G is introduced, which supports classification problems for predefined non ordered categories by a group of decision makers. In NeXClass-G aggregation of individual preferences is executed at the parameter level, which is used as input set for NeXClass. For the aggregation of preferences an operator based on Social Judgment Scheme is introduced, in order to maximize consent. Moreover, for group acceptance on parameters aggregation of opinions is introduced based on WOWA operator with appropriate weight definition, in order to express the majority concept. Satisfaction and acceptance level are calculated by definition of consent indices. Finally, the prototype Group Decision Support Systems (GDSS) NeXClass-GDSS is presented, which supports group decisions in multicriteria classification problems for predefined non ordered categories. Both methodologies (for single decision maker and group), as well as the GDSS have been applied in real world classification problems in banking domain and evaluated. Findings provide strong evidence that methodologies as well as GDSS sufficiently cover the needs and contribute in a positive and prototype way in the field of multicriteria analysis and group decisions in alignment with dissertations' research goals.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2008
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Rigopoulos, G.
- Ρηγόπουλος, Γεώργιος
Subjects
dc:subject × 18- Πολυκριτήρια ανάλυση
- Ομαδικές αποφάσεις
- Ταξινόμηση
- Μη διατεταγμένες κατηγορίες
- ΣΥΑ
- ΣΥΟΑ
- Τελεστές σύνθεσης
- OWA operators
- WOWA
- Multicriteria analysis
- Group decision making
- Classification
- Decision support systems
- Aggregation operators
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/17008