Inferring an ELECTRE TRI Model from Assignment Examples

  • Authors:
  • V. Mousseau;R. Slowinski

  • Affiliations:
  • LAMSADE, Université Paris Dauphine, Place du Maréchal De Lattre de Tassigny, 75775 Paris Cedex 16, France. Tel: (33-1)44-05-44-01, fax: (33-1)44-05-40-91. E-mail: mousseau@lamsade.dauphi ...;Institute of Computing Science, Poznan University of Technology, Piotrowo 3a, 60-965 Poznan, Poland, tel: 48-61-790790, fax: 48-61-771525 (Email: slowinsk@sol.put.poznan.pl)

  • Venue:
  • Journal of Global Optimization
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

Given a finite set of alternatives, the sorting problem consists in theassignment of each alternative to one of the pre-defined categories. In thispaper, we are interested in multiple criteria sorting problems and, moreprecisely, in the existing method ELECTRE TRI. This method requires theelicitation of parameters (weights, thresholds, category limits,...) inorder to construct the Decision Maker‘s (DM) preference model. A directelicitation of these parameters being rather difficult, we proceed to solvethis problem in a way that requires from the DM much less cognitive effort.We elicit these parameters indirectly using holistic information given bythe DM through assignment examples. We propose an interactive approach thatinfers the parameters of an ELECTRE TRI model from assignment examples. Thedetermination of an ELECTRE TRI model that best restitutes the assignmentexamples is formulated through an optimization problem. The interactiveaspect of this approach lies in the possibility given to the DM to revisehis/her assignment examples and/or to give additional information before theoptimization phase restarts.