<b Pruning the tree:
Classification and regression trees, applied to data with known values for an outcome variable, derive models with rules like "If taxable income <$80,000, if no Schedule C income, if standard deduction taken, then no-audit." Pruning is the process of truncating the rules (= pruning the branches of the tree) so they are less complex. This corrects for the tendency of the tree to overfit the model to the noise in the data it was trained on, reducing its accuracy with new data.