Version: | 1.0 |
---|---|
Provider: | R-MCDA |
Name: | MRSort_inference_approx |
SOAP service’s name: | |
MRSort_inference_approx-R-MCDA (see SOAP requests for details) |
MRSort is a simplification of the Electre TRI method that uses the pessimistic assignment rule, without indifference or preference thresholds attached to criteria. Only a binary discordance condition is considered, i.e. a veto forbids an outranking in any possible concordance situation, or not. The identification of the profiles, weights, majority threshold and veto thresholds are done by taking into account assignment examples.
Contact: Alexandru Olteanu (alexandru.olteanu@univ-ubs.fr)
(For outputs, see below)
A complete list of alternatives to be considered when inferring the MR-Sort model.
The input value should be a valid XMCDA document whose main tag is <alternatives>
.
The evaluations of the alternatives on the set of criteria.
The input value should be a valid XMCDA document whose main tag is <performanceTable>
.
A list of criteria on which the alternatives are evaluated. For each criterion, the preference direction should be provided.
The input value should be a valid XMCDA document whose main tag is <criteriaScales>
.
The alternatives assignments to categories.
The input value should be a valid XMCDA document whose main tag is <alternativesAssignments>
.
A list of categories ranks, 1 stands for the most preferred category and the higher the number the lower the preference for that category.
The input value should be a valid XMCDA document whose main tag is <categoriesValues>
.
The method parameters.
The input value should be a valid XMCDA document whose main tag is <programParameters>
.
It must have the following form:
<programParameters>
<parameter id="veto">
<values>
<value>
<boolean>%1</boolean>
</value>
</values>
</parameter>
<parameter id="time">
<values>
<value>
<integer>%2</integer>
</value>
</values>
</parameter>
<parameter id="population">
<values>
<value>
<integer>%3</integer>
</value>
</values>
</parameter>
<parameter id="mutation">
<values>
<value>
<real>%4</real>
</value>
</values>
</parameter>
<parameter id="test">
<values>
<value>
<boolean>%5</boolean>
</value>
</values>
</parameter>
</programParameters>
where:
%1 is a parameter named “Include vetoes”. This is a boolean. The default value is false.
%2 is a parameter named “Time limit”. This is a int, and the value should conform to the following constraint: An integer value (minimum 1). More formally, the constraint is:
%2 > 0
The default value is 60.
%3 is a parameter named “Population size”. This is a int, and the value should conform to the following constraint: An integer value (minimum 10). More formally, the constraint is:
%3 >= 10
The default value is 20.
%4 is a parameter named “Mutation probability”. This is a float, and the value should conform to the following constraint: A value between 0 and 1. More formally, the constraint is:
%4 >= 0 && %4 <= 1
The default value is 0.1.
%5 is a parameter named “Test”. This is a boolean. The default value is false.
The evaluations of the category profiles.
The returned value is a XMCDA document whose main tag is <performanceTable>
.
The evaluations of the veto profiles.
The returned value is a XMCDA document whose main tag is <performanceTable>
.
The criteria weights.
The returned value is a XMCDA document whose main tag is <criteriaValues>
.
The categories delimiting profiles.
The returned value is a XMCDA document whose main tag is <categoriesProfiles>
.
The categories veto profiles.
The returned value is a XMCDA document whose main tag is <categoriesProfiles>
.
The model fitness.
The returned value is a XMCDA document whose main tag is <programParameters>
.
The majority threshold.
The returned value is a XMCDA document whose main tag is <programParameters>
.
Messages from the execution of the webservice. Possible errors in the input data will be given here.
The returned value is a XMCDA document whose main tag is <programExecutionResult>
.