EURO 2025 Leeds
Abstract Submission

2551. Modelling Confidence and Optimism in Preference Learning

Invited abstract in session MC-8: Preference Learning 2, stream Multiple Criteria Decision Aiding.

Monday, 12:30-14:00
Room: Clarendon SR 2.08

Authors (first author is the speaker)

1. Sally Giuseppe Arcidiacono
Department of Economics and Business, University of Catania
2. Salvatore Corrente
Department of Economics and Business, University of Catania
3. Salvatore Greco
Department of Economics and Business, University of Catania

Abstract

In this work, we explore a recently proposed model of ordinal regression within multicriteria decision analysis, which assumes that the decision maker considers the entire set of feasible weight vectors. The overall evaluation of each alternative is obtained through a non-additive average of the evaluations provided by each weight vector. This average is computed by considering a non-additive probability in the weight vector space and aggregating the evaluations with respect to each weight vector using the Choquet integral. The non-additive probability is derived as a transformation of an additive probability. The dispersion around a reference vector of the additive probability measures the decision maker’s confidence, while the transformation itself represents the optimism, with higher values indicating greater optimism. We evaluate the ordinal regression methodology based on confidence and optimism across several datasets, discussing the performance and potential of the proposed approach.

Keywords

Status: accepted


Back to the list of papers