Decision Tree (DT)

Pros:

• Easy to interpret

• You don’t have to worry about outliers Cons:

• Time-consuming process

• Decision Trees (DT) are highly sensitive to training data

• Tend to overfit http://blog.echen.me/2011/04/27/choosing-a-machine-learning-classifier/

• Gini Index can be used to determine purity and when to split nodes

• Information Gain vs GINI Index

Types:

• Conditional Inference Trees

Classifiers:

• ID3 page=41
• C4.5 Algorithm