Random Sample Options

Oracle Data Miner supports the following types of sampling:

A random sample is created by choosing cases at random, that is, in such a way that each item in the data set has the same chance of being selected. The random sample is smaller than the set that it is based on. Using a smaller data set can result in improved performance of model build and apply. If the sample is representative of the whole set, results generated using the sample will be compatible to the results generated using the whole set.

The Total number of cases is displayed in a read-only field.

You can specify:

Click OK to close the dialog and save any changes.