When selecting the best combination of feature and value as the splitting point, two criteria such as Gini Impurity and Information Gain can be used to measure the quality of separation.
Gini Impurity, as its name implies, measures the impurity rate of the class distribution of data points, or the class mixture rate. For a dataset with K classes, suppose data from class 
take up a fraction 
of the entire dataset, the Gini Impurity of this dataset is written as follows:
Lower Gini Impurity indicates a purer ...
