A stopping criterion tells a decision tree when to stop splitting a node. It limits overfitting by preventing leaves from becoming too small or overly specialized. section 9.2
Common stopping criteria include:
- Stop when the improvement in accuracy, Gini index, or entropy falls below a threshold. section 9.2
- Stop when a node has fewer than a specified number of samples. section 9.2
- Split only when both resulting leaves contain enough samples. section 9.2
- Stop after reaching a maximum tree depth. section 9.2
App recommendations
A tree can first split users by age, then split the adult group by platform to create pure leaves. A stopping criterion can prevent further splits once leaves are sufficiently small or pure. section 9.2
Table 9.2 A simplified version of the dataset in table 9.1, where the age column has been simplified to two categories, “young” and “adult” (view table figure)
| Platform |
Age |
App |
| iPhone |
Young |
Atom Count |
| iPhone |
Adult |
Check Mate Mate |
| Android |
Adult |
Beehive Finder |
| iPhone |
Adult |
Check Mate Mate |
| Android |
Young |
Atom Count |
| Android |
Young |
Atom Count |
Use the simplified user dataset to trace how candidate splits create leaves and where a stopping criterion could limit further growth. section 9.2
