Computing Gini index
The decision tree algorithm aims to achieve partitions in the terminal nodes that are as pure as possible. The Gini index is one of the methods used to achieve this. It is calculated based on the proportion of samples in each group.
Given the number of people who stayed and left respectively, calculate the Gini index for that node.
Deze oefening maakt deel uit van de cursus
HR Analytics: Predicting Employee Churn in Python
Oefeninstructies
- Calculate the total number of employees in that node.
- Compute the Gini index based on proportion of employees in each group.
Praktische interactieve oefening
Probeer deze oefening eens door deze voorbeeldcode in te vullen.
#number of people who stayed/left
stayed = 37
left = 1138
#sum of stayed and left
total = ____ + ____
#gini index
gini = ____*(____/total)*(____/total)