You are here: Symbol Reference > DBClassifier Namespace > Classes > TDBClassifier Class > TDBClassifier Methods > TDBClassifier.ClassifyTest Method
Data Miner
TDBClassifier.ClassifyTest Method

Perform classification test.

function ClassifyTest(Split: integer = 0): double;

The method calls OnClassifyDataSet event and passes to it the Split parameter. When performing classification tests, the given dataset is split in to two sets: the learn set and the test set. To obtain a more reliable estimate, the classification test is performed on a set of splits. (30 for example). The Split parameter should therefore provide a way to define the split on which the classification test is to be run. The accuracy of the classification is returned as the result of the function. If there are 30 splits the ClassifyTeset function should be called 30 times and the results should then be statistically analyzed.

Copyright (c) 1999-2017 by Dew Research. All rights reserved.
What do you think about this topic? Send feedback!