RORUTADIS-StochasticResults

Version:0.1
Provider:PUT
SOAP service’s name:
 RORUTADIS-StochasticResults-PUT (see SOAP requests for details)

Description

Calculates stochastic results for alternative assignments, assignment-based preference relation and class cardinalities. The results are computed by sampling the space of compatible models.

Contact:
Krzysztof Ciomek (k.ciomek@gmail.com), Milosz Kadzinski (milosz.kadzinski@cs.put.poznan.pl)

Web page: https://github.com/kciomek/rorutadis

Reference: None

Inputs

(For outputs, see below)

criteria

A list of criteria (<criteria> tag) with information about preference direction (<criteriaValues mcdaConcept=”preferenceDirection”>, 0 - gain, 1 - cost) and number of characteristic points (<criteriaValues mcdaConcept=”numberOfCharacteristicPoints”>, 0 for the most general marginal utility function or integer grater or equal to 2) of each criterion.

The input value should be a valid XMCDA document whose main tag is <criteria>. It must have the following form:

<criteria>
        <criterion id="[...]" />
        [...]
</criteria>

<criteriaValues mcdaConcept="preferenceDirection">
        <criterionValue>
                <criterionID>[...]</criterionID>
                <value><integer>[...]</integer></value>
        </criterionValue>
        [...]
</criteriaValues>

<criteriaValues mcdaConcept="numberOfCharacteristicPoints">
        <criterionValue>
                <criterionID>[...]</criterionID>
                <value><integer>[0|integer greater or equal to 2]</integer></value>
        </criterionValue>
        [...]
</criteriaValues>

alternatives

A list of alternatives.

The input value should be a valid XMCDA document whose main tag is <alternatives>. It must have the following form:

                    <alternatives>
    <alternative id="[...]">
        <active>[...]</active>
    </alternative>
    [...]
</alternatives>

categories

A list of categories (classes). List must be sorted from the worst category to the best.

The input value should be a valid XMCDA document whose main tag is <categories>. It must have the following form:

                    <categories>
    <category id="[...]" />
    [...]
</categories>

performanceTable

The performances of the alternatives.

The input value should be a valid XMCDA document whose main tag is <performanceTable>.


assignmentExamples (optional)

A list of assignment examples of alternatives to intervals of categories (classes) or to a specific category (class).

The input value should be a valid XMCDA document whose main tag is <alternativesAffectations>. It must have the following form:

<alternativesAffectations>
        <alternativeAffectation>
                <alternativeID>[...]</alternativeID>
                <categoryID>[...]</categoryID>
        </alternativeAffectation>
        [...]
        <alternativeAffectation>
                <alternativeID>[...]</alternativeID>
                <categoriesInterval>
                        <lowerBound>
                                <categoryID>[...]</categoryID>
                        </lowerBound>
                        <upperBound>
                                <categoryID>[...]</categoryID>
                        </upperBound>
                </categoriesInterval>
        </alternativeAffectation>
        [...]
        <alternativeAffectation>
                <alternativeID>[...]</alternativeID>
                <categoriesSet>
                        <categoryID>[...]</categoryID>
                        [...]
                </categoriesSet>
        </alternativeAffectation>
        [...]
</alternativesAffectations>

assignmentComparisons (optional)

Two lists of assignment pairwise comparisons. A comparison from list with attribute mcdaConcept=”atLeastAsGoodAs” indicates that some alternative should be assigned to class at least as good as class of some other alternative (k = 0) or at least better by k classes (k > 0). A comparison from list with attribute mcdaConcept=”atMostAsGoodAs” indicates that some alternative should be assigned to class at most better by k classes (k > 0) then some other alternative. Note: usage of this kind of preference information significantly slows down computations.

The input value should be a valid XMCDA document whose main tag is <alternativesComparisons>. It must have the following form:

<alternativesComparisons mcdaConcept="atLeastAsGoodAs">
        <pairs>
                <pair>
                        <initial><alternativeID>[...]</alternativeID></initial>
                        <terminal><alternativeID>[...]</alternativeID></terminal>
                        <value><integer>k</integer></value>
                </pair>
                [...]
        </pairs>
</alternativesComparisons>

<alternativesComparisons mcdaConcept="atMostAsGoodAs">
        <pairs>
                [...]
        </pairs>
</alternativesComparisons>

categoriesCardinalities (optional)

A list of category (class) cardinality constraints. It allows to define minimal and/or maximal desired category (class) cardinalities. Note: usage of this kind of preference information significantly slows down computations.

The input value should be a valid XMCDA document whose main tag is <categoriesValues>. It must have the following form:

<categoriesValues>
        <categoryValue>
                <categoryID>[...]</categoryID>
                <value>
                        <interval>
                                <lowerBound><integer>[...]</integer></lowerBound>
                                <upperBound><integer>[...]</integer></upperBound>
                        </interval>
                </value>
        </categoryValue>
        [...]
</categoriesValues>

methodParameters

Method parameters.

The input value should be a valid XMCDA document whose main tag is <methodParameters>. It must have the following form:

<methodParameters>
    <parameter name="strictlyMonotonicValueFunctions">
        <value>
            <boolean>%1</boolean>
        </value>
    </parameter>
    <parameter name="numberOfSamples">
        <value>
            <integer>%2</integer>
        </value>
    </parameter>
<!– The following is OPTIONAL - we comment it out because we do not want to
change the WS nor the diviz program for now, we just want to be able to use it in automated tests.
–><!–
<parameter name=”seed”>
<value>
<integer>%3</integer>

</value>

</parameter>

–>
</methodParameters>

where:

  • %1 is a parameter named “strictlyMonotonicValueFunctions”. This is a boolean. The default value is false.
  • %2 is a parameter named “numberOfSamples”. This is a int. The default value is 100.

Outputs

stochasticAssignments

Stochastic assignments. The value for alternative a_i and category c_j equals to the rate of samples, for which alternative a_i was assigned to category C_j.

The returned value is a XMCDA document whose main tag is <alternativesAffectations>. It has the following form:

<alternativesAffectations>
        <alternativeAffectation>
                <alternativeID>[...]</alternativeID>
                <categoryID>[...]</categoryID>
                <value>
                        <real>[...]</real>
                </value>
        </alternativeAffectation>
        [...]
</alternativesAffectations>

stochasticPreferenceRelation

Stochastic preference relation. Each value for pair (a_i, a_j) describes the rate of samples, for which alternative a_i was assigned to class at least as good as class of a_j.

The returned value is a XMCDA document whose main tag is <alternativesComparisons>. It has the following form:

<alternativesComparisons mcdaConcept="atLeastAsGoodAs">
        <pairs>
                <pair>
                        <initial>
                                <alternativeID>[...]</alternativeID>
                        </initial>
                        <terminal>
                                <alternativeID>[...]</alternativeID>
                        </terminal>
                        <value>
                                <real>[...]</real>
                        </value>
                </pair>
                [...]
        </pairs>
</alternativesComparisons>

stochasticClassCardinalities

Stochastic class cardinalities. Each category has sequence of n + 1 values, where n is the number of considered alternatives. First value corresponds to the cardinality 0, and next n values to subsequent cardinalities from 1 to n. i-th value for c_j class describes the rate of samples, for which i-1 alternatives were assigned to class c_j.

The returned value is a XMCDA document whose main tag is <categoriesValues>. It has the following form:

<categoriesValues>
        <categoryValue>
                <categoryID>c01</categoryID>
                <values>
                        <value>
                                <real>[...]</real>
                        </value>
                        <value>
                                <real>[...]</real>
                        </value>
                        [...]
                </values>
        </categoryValue>
        [...]
</categoriesValues>

messages

Messages generated by the program.

The returned value is a XMCDA document whose main tag is <methodMessages>.


Original xml description