Data Science Essentials

An interdisciplinary field using statistics, programming, and domain knowledge to extract insights from data.

Problem definition → data collection → cleaning → exploration → modeling → evaluation → deployment.

Cross-Industry Standard Process for Data Mining — a structured methodology.

A collection of related data, typically organized in rows and columns.

A measurable property or characteristic used as input to a model.

The output a model tries to predict.

Learning from labeled examples to predict outcomes.

Finding patterns in unlabeled data.

Learning by taking actions and receiving rewards/penalties.

Learning from a mix of labeled and unlabeled data.

Predicting a continuous numeric output.

Predicting a categorical output.