... split?–Determine when to stop splittingData Mining Classification: Basic Concepts, Decision Trees, and Model EvaluationLecture Notes for Chapter 4Introduction to Data MiningbyTan, Steinbach, Kumar© ... NoNode N1 Node N2Before Splitting:M0M1M2 M3 M4M12M34Gain = M0 – M12 vs M0 – M34© Tan,Steinbach, Kumar Introduction to Data Mining 31 Measure of Impurity: GINIGini Index for a given node ... among all classes, implying least interesting information–Minimum (0.0) when all records belong to one class, implying most interesting information∑−=jtjptGINI2)]|([1)(C1 0C2 6Gini=0.000C1...