PROMETHEE-Cluster

Provider:PUT
Version:1.0

Description

Clustering using Promethee Tri and k-means algorithm

Inputs

(For outputs, see below)

criteria

Description:

Information about criterias required to compute Promethee Tri: criteria id, criteria thresholds (indifference and preference) and criteria scales.

alternatives

Description:

List of alternatives.

XMCDA related:

  • Tag: alternatives

performance table

Description:

The performance of alternatives.

XMCDA related:

  • Tag: performanceTable

weights

Description:

Weights of criteria to consider.

XMCDA related:

  • Tag: criteriaValues

criteria scales

Description:

None

XMCDA related:

  • Tag: criteriaScales

criteria thresholds

Description:

None

XMCDA related:

  • Tag: criteriaThresholds

parameters

Description:

First parameter specifies number of output clusters K.
Where K is integer greater than 1 and less than number of alternatives. Second parameter specifies parameter P of deviation used in Promethee TRI. In this case deviation is weighted generalised mean of difference between flows on criteria, where P is used exponent.

GUI information:

  • Parameter values can be defined via the GUI or the XMCDA file, by default via GUI.
  • Name: number of clusters
    • Constraint description: A integer value greater than 1 and lower than the number of alternatives
    • Type: integer
  • Name: distance
    • Type: integer
    • Default value: 1
  • Name: Use random seed
    • Default value: false
  • Name: random seed
    • Type: integer

XMCDA related:

  • Tag: programParameters

  • Code:

    <programParameters>
    %1
    %2
<!– %3 –>
%4

</programParameters>

Outputs

alternatives sets

Description:

List of alternatives sets where each set representing one cluster.

XMCDA related:

  • Tag: alternativesSets

messages

Description:

Messages or errors generated by this module.

XMCDA related:

  • Tag: programExecutionResult

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