Learning from Observations

Outline

What is Learning

Why is it necessary?

Learning agents

Learning element

Induction

Inductive learning

Inductive learning method

Inductive learning method

Inductive learning method

Inductive learning method

Inductive learning method

Inductive learning method

Inductive learning method

An application: Ad blocking

Learning Ad blocking

Nearest Neighbor

Application: Eating out

Attribute representation

Bayes Rule

Naïve Bayes Classifier

Naïve Bayes Algorithm

An Example

An Example

Decision trees

Expressiveness

Hypothesis spaces

Hypothesis spaces

The best hypothesis

Decision tree learning

Choosing an attribute

Information Content

Entropy curve

Using information theory

Information gain

Information gain

Example contd.

Performance measurement

Training and testing sets

Overfitting

Summary