Data Generator
Creates random data containing some clusters for Parallel Universes.
The data contains a certain fraction of noise patterns and data that
is generated to clusters (all clusters have the same size). The data is
normalized in [0, 1].
Dialog Options
- Cluster Count
-
The number of clusters for each universe (comma separated).
- Universe Sizes
-
The number of attributes for each universe (comma separated).
- Pattern Count
-
Number of patterns (overall).
- Standard Deviation
-
Std. dev. for each cluster.
- Noise Fraction
-
Fraction of the overall data to be noise, i.e. randomly drawn data.
This value must be in [0, 1].
- Random Seed
-
Seed for "deterministic" data generation.
Ports
Output Ports
0 |
Contains the data with the cluster id as last column |
1 |
Contains the cluster centers. The attributes in the universes where
the cluster is not located, are filled with missing values. |
This node is contained in KNIME Base Nodes
provided by KNIME GmbH, Konstanz, Germany.