LearnMRSortMeta

Provider:oso
Version:1.0

Description

The purpose of this webservice is to infer parameters of an ELECTRE TRI Bouyssou-Marchant model, also called MR-Sort model. It takes in input the categories and criteria of the model and examples of assignments. The user should also specify several parameters of the metaheuristic: the size of the population of models (nmodels), the number of iterations to do with the heuristic adjusting the profiles, and the number of iterations (niter_heur) of the main loop (niter_meta). The metaheuristic will try to find a MRSort model which is compatible with the highest number of examples given in input.

  • Contact: Olivier Sobrie (olivier.sobrie@gmail.com)
  • Reference: Olivier Sobrie, Vincent Mousseau, Marc Pirlot: Learning a Majority Rule Model from Large Sets of Assignment Examples

Inputs

(For outputs, see below)

criteria

Description:

The list of criteria of the model.

alternatives

Description:

The list of learning alternatives to use to learn the MR-Sort model.

XMCDA related:

  • Tag: alternatives

categories

Description:

The list of categories of the model and their rank

XMCDA related:

  • Tag: categories

performanceTable

Description:

The performance table of the learning alternatives.

XMCDA related:

  • Tag: performanceTable

alternativesAssignments

Description:

The assignments of the alternatives in the different categories.

XMCDA related:

  • Tag: alternativesAffectations

params

Description:

The parameters of the metaheuristic. ‘nmodels’: the size of the population; ‘niter_heur’: the number of iterations of the heuristic; ‘niter_meta’: the number of iterations of the main loop.

XMCDA related:

  • Tag: methodParameters

solver

Optional: yes, enabled by default

Description:

The solver to use. Currently CPLEX and GLPK are supported. In none specified, GLPK is used.

XMCDA related:

  • Tag: methodParameters

Outputs

compatibleAlternatives

Description:

The reference alternatives that are compatible with the profiles, weights and majority threshold computed by the webservice.

XMCDA related:

  • Tag: alternatives

profilesPerformances

Description:

The profiles performance table of the profiles computed by the webservice.

XMCDA related:

  • Tag: performanceTable

criteriaWeights

Description:

The set of criteria weights found by the webservice.

XMCDA related:

  • Tag: criteriaValues

categoriesProfiles

Description:

The category profiles computed by the webservice.

XMCDA related:

  • Tag: categoriesProfiles

majorityThreshold

Description:

The majority threshold computed by the webservice.

XMCDA related:

  • Tag: methodParameters

messages

Description:

A list of messages generated by the webservice. In this output the result of the inference will be given. It gives informations on what might be wrong in the inputs.

XMCDA related:

  • Tag: methodMessages

For further technical details on the web service underlying this program, have a look at its documentation here.