Desktop Survival Guide
by Graham Williams
Validation Set: A portion of a dataset that is used to test the performance of a model as it is being built. See also Training Set and Test Set.
Variance: The error in a model due to the fact that the model is built from only a sample of the data taken to represent the whole population. That is, those instances occasionally incorrectly classified by models built by the learning algorithm. Modelling error due to variance can be reduced using Bagging. Compare with bias.