A.M.L. - Chapter 3: Decision Tree Classifier

What is a decision tree classifier?

A supervised learning model used for classification and regression that predicts outcomes by following decision paths from root to leaf nodes.

What are the types of nodes in a decision tree?

Root node, internal nodes, and leaf nodes.

What is the goal of decision tree induction?

To create a model that predicts the output variable based on input variables using historical data.

What is entropy in decision trees?

A measure of impurity or randomness in the dataset. Lower entropy means higher purity.

How is information gain used in decision trees?

It helps select the attribute that best separates the classes by maximizing the reduction in entropy.

What is pruning in decision trees?

The process of removing unnecessary branches to reduce model complexity and prevent overfitting.

Why are smaller decision trees preferred?

They are more interpretable, generalize better, and reduce the risk of overfitting.

How can a decision tree be converted to rules?

Each path from root to leaf becomes an "if-then" rule describing a classification decision.

What is rule pruning?

Removing non-critical conditions or rules to simplify rule sets and improve generalization.

What Python library and class are used to build decision trees?

Scikit-learn’s DecisionTreeClassifier from the tree module.

What is a real-world advantage of using decision trees?

They can operate without full data upfront, making them ideal for applications like medical diagnosis.