Synopsis
Delivers a performance based on cluster centroids.
Description
An evaluator for centroid based clustering methods. The average within cluster distance is calculated by averaging the distance between the centroid and all examples of a cluster.
Input
- example set: expects: ExampleSet
- cluster model: expects: CentroidClusterModel
- performance: optional: PerformanceVector
Output
- performance:
- example set:
- cluster model:
Parameters
- main criterion: The main criterion to use
- main criterion only: return the main criterion only
- normalize: divide the criterion by the number of features
- maximize: do not multiply the result by minus one