Netezza Divisive Clustering Build Options

The Build Options tab is where you set all the options for building the model. You can, of course, just click the Run button to build a model with all the default options, but normally you will want to customize the build for your own purposes.

Distance measure. The method to be used for measuring the distance between data points; greater distances indicate greater dissimilarities. The options are:

• Euclidean. (default) The distance between two points is computed by joining them with a straight line.
• Manhattan. The distance between two points is calculated as the sum of the absolute differences between their co-ordinates.
• Canberra. Similar to Manhattan distance, but more sensitive to data points closer to the origin.
• Maximum. The distance between two points is calculated as the greatest of their differences along any coordinate dimension.

Maximum number of iterations. The algorithm operates by performing several iterations of the same process. This option allows you to stop model training after the number of iterations specified.

Maximum depth of cluster trees. The maximum number of levels to which the data set can be subdivided.

Replicate results. Check this box if you want to set a random seed, which will enable you to replicate analyses. You can either specify an integer or click Generate, which creates a pseudo-random integer.

Minimum number of instances for a split. The minimum number of records that can be split. When fewer than this number of unsplit records remain, no further splits will be made. You can use this field to prevent the creation of very small subgroups in the cluster tree.