Szczegóły publikacji
Opis bibliograficzny
Reconciling inconsistent preference information in group multicriteria decision support with reference sets / Andrzej M. J. SKULIMOWSKI // W: Advances and trends in artificial intelligence : theory and applications : 36th international conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2023 : Shanghai, China, July 19–22, 2023 : proceedings, Pt. 1 / eds. Hamido Fujita [et al.]. — Cham : Springer Nature, cop. 2023. — (Lecture Notes in Computer Science ; ISSN 0302-9743 ; LNCS 13925. Lecture Notes in Artificial Intelligence). — ISBN: 978-3-031-36818-9; e-ISBN: 978-3-031-36819-6. — S. 207–220. — Bibliogr., Abstr. — Publikacja dostępna online od: 2023-07-19. — A. M. J. Skulimowski - dod. afiliacja: International Centre for Decision Sciences and Forecasting, Progress and Business Foundation, Kraków
Autor
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 147843 |
|---|---|
| Data dodania do BaDAP | 2023-09-14 |
| DOI | 10.1007/978-3-031-36819-6_18 |
| Rok publikacji | 2023 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Wydawca | Springer |
| Konferencja | International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems 2023 |
| Czasopismo/seria | Lecture Notes in Computer Science |
Abstract
Home Advances and Trends in Artificial Intelligence. Theory and Applications Conference paper Reconciling Inconsistent Preference Information in Group Multicriteria Decision Support with Reference Sets Andrzej M. J. Skulimowski Conference paper First Online: 19 July 2023 Part of the Lecture Notes in Computer Science book series (LNAI,volume 13925) Abstract This article proposes an algorithm to reconcile inconsistent recommendations of experts involved in a multicriteria decision support procedure. The algorithm yields a consistent set of reference points transformed from an arbitrary collection. By assumption, experts S1,…,Sn are agents independently involved in a decision making process of other agents termed decision makers. Experts formulate recommendations as reference points in the criteria space and simultaneously communicate them to decision makers. The above decision-making process corresponds to the schemes ‘one decision maker – multiple recommending experts’ (group decision support) or ‘multiple decision makers – multiple experts’ (group decision making and support). The assumed independence of expert judgments may provide different types of recommendation inconsistency. We define several variants of the internal and mutual inconsistency, which may occur simultaneously in the same decision-making problem. We will also assume that expert recommendations may belong to four predefined characteristic reference sets. The proposed new preference aggregation procedure regularizes the set of reference values pointed out by multiple experts. The aggregation-regularization operations include recommendation merging, reference point averaging, splitting of reference classes, moving a reference point between classes, or removing it from a class. A real-life example displaying the implementation of content-based multimedia retrieval from a knowledge repository will illustrate the above approach. In the final section, we will discuss the dependence of aggregation-regularization process outcomes on the sequence of reference classes, with reference points within each class checked for inconsistencies.
