Inferring Preferences for Multi-Criteria Ordinal Classification Methods Using Evolutionary Algorithms

Multicriteria sorting involves assigning the objects of decisions (actions) into $a$ priori known ordered classes considering preferences a decision maker (DM). Two new multicriteria methods were recently proposed by authors. These are based on novel approach called interval-based outranking which provides with attractive practical and theoretical characteristics. However, as is well known, defining parameter values for often very difficult. This difficulty arises not only from large number parameters DM’s lack familiarity them, but also imperfectly (even missing) information. Here, we address: i) how to elicit two methods, ii) incorporate imperfect knowledge during elicitation. We follow preference disaggregation paradigm use evolutionary algorithms address it. Our proposal performs in wide range computational experiments. Interesting findings are: method restores assignment examples high effectiveness using three profiles each limiting boundary or representative actions per class; ability appropriately assign unknown can be greatly improved increasing profiles.