Szczegóły publikacji
Opis bibliograficzny
Fusion of expert information in content-based multimedia retrieval with group decision support / Andrzej M. J. SKULIMOWSKI // W: PP-RAI'2022 [Dokument elektroniczny] : proceedings of the 3rd Polish conference on Artificial intelligence : April 25–27, 2022, Gdynia, Poland. — Wersja do Windows. — Dane tekstowe. — Gdynia : Gdynia Maritime University, 2022. — e-ISBN: 978-83-7421-401-8. — S. 159–163. — Wymagania systemowe: Adobe Reader. — Tryb dostępu: https://wydawnictwo.umg.edu.pl/pp-rai2022/pdfs/ProceedingsPP-... [2022-04-27]. — Bibliogr. s. 163, Abstr. — Dod. afiliacja autora: International Centre for Decision Sciences and Forecasting, Progress & Business Foundation, Kraków
Autor
Słowa kluczowe
Dane bibliometryczne
ID BaDAP | 140001 |
---|---|
Data dodania do BaDAP | 2022-04-29 |
Rok publikacji | 2022 |
Typ publikacji | materiały konferencyjne (aut.) |
Otwarty dostęp | |
Wydawca | Uniwersytet Morski w Gdyni |
Abstract
This paper proposes a new approach to fuse recommendations of experts involved in multicriteria decision support procedures. Recommendations are formulated by experts independently as elements of four predefined reference sets. The decision situation corresponds either to the scheme ‘one decision maker – multiple recommending experts’ (group decision support) or ‘multiple decision makers – multiple experts’ (group decision making and support). We define the internal, mutual, and plausibility inconsistencies and propose a procedure to verify consistency, regularize the reference values, and fuse them. An expert’s trust coefficient decrease when a reference value recommended by this expert is modified during the regularization process. Thus, the regularization outcomes can be assessed ex ante with respect to the total losses of experts’ credibility and to the quality of the resulting reference set. The decision process will be illustrated by an example of content-based multimedia retrieval from the AI-based knowledge repository developed within a recent research project.