K Means
This task create k clusters based upon the specified feature or features.
CONFIGURATION
| OPTION | DESCRIPTION |
|---|---|
| File Name | Use the browse button to select the CSV file which is to be read. The effective file name represents the interpolated and fully qualified file name. This is useful if there are environment variable components to the path. |
| Limit Number Of Rows | When checked, this will impose a maximum number of rows to read. This is useful for testing large datasets with a small initial sample. |
INPUT
None.
OUTPUT
The CSV file will be output to the Dex data stream. The next component in the flow will receive this as input.
Sample Visualization
Here is a view of the performance of our age cluster. The first column is age, the 2nd one represents cluster and the third column represents whether the individual survived (1) or perished (0). Using this, we can visualize the performance of this cluster relative to survival.
