Cluster Statistics
The Statistics table in the Cluster Statistics section gives a quick overview of key information.
The following information is provided:
- The names of the clusters
- The absolute sizes of the clusters
- The sizes of the clusters in relation to the total number of records. For example, a value of 20% in the Size column means that 20% of all records belong to this cluster.
- The homogeneity of the clusters. The homogeneity factor indicates how similar the records are that belong to one cluster. This column is only available if you have selected the Show algorithm-specific information check box on the General page of the Properties notebook, and if this information is available in the model that you are viewing.
The
Similarity Between Clusters table shows the similarity between the
clusters. Each cluster is compared to all other clusters. The similarity
value
must be between 0.0 and 1.0.
- 0.0 means that the clusters are completely different.
- 1.0 means that the clusters are identical.
If the field is not numeric, or if the
information is not available in
the statistics of the model, the following fields do not contain any
values.
This is denoted by the string n/a (not available).
- Minimum
- Maximum
- Mean
- Standard deviation
- Distance unit
- Aggregated value