When we remove sub-nodes of a Decision node, this process is called pruning or the opposite process of splitting. The two techniques which are widely used for pruning are- Post and Pre Pruning.
Post Pruning:
- This type of pruning is used after the construction of the Decision Tree.
- This technique is used when the Decision Tree will have a very large depth and will show the overfitting of the model.
- It is also known as backward pruning.
- This technique is used when we have an infinitely grown Decision Tree.
Pre Pruning:
- This technique is used before the construction of the Decision Tree.
- Pre-Pruning can be done using Hyperparameter tuning.
- Overcome the overfitting issue.