On the learning of weights in some aggregation operators: the weighted mean and OWA operators

We study the determination of weights for two types of aggregation operators: the weighted mean and the OWA operator. We assume that there is at our disposal a set of examples for which the outcome of the aggregation operator is known. In the case of the OWA operator, we compare the results obtained...

ver descrição completa

Detalhes bibliográficos
Autor: Torra Ferré, Vicenç
Formato: artículo
Fecha de publicación:1999
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2099/3557
Acesso em linha:https://hdl.handle.net/2099/3557
Access Level:acceso abierto
Palavra-chave:Learning weigths
Modelling
Aggregation operators
Weighted mean
OWA operators
WOWA operators
Conjunts, Teoria de
Classificació AMS::03 Mathematical logic and foundations::03E Set theory
Descrição
Resumo:We study the determination of weights for two types of aggregation operators: the weighted mean and the OWA operator. We assume that there is at our disposal a set of examples for which the outcome of the aggregation operator is known. In the case of the OWA operator, we compare the results obtained by our method with another one in the literature. We show that the optimal weighting vector is reached with less cost.