7 Introduction to Machine Learning Free Online FlashCards

What is machine learning?

Machine learning uses data to learn patterns that help a program make predictions or organize information.

What distinguishes supervised learning?

Supervised learning uses examples that pair input features with a desired answer, or label, to learn predictions for new examples.

What does regression predict?

Regression predicts a numeric value, such as a home's sale price or tomorrow's temperature.

What does classification predict?

Classification predicts a category, such as whether an email is spam or which plant species appears in a photo.

What distinguishes unsupervised learning?

Unsupervised learning finds structure in examples that have features but no supplied target labels.

What is clustering?

Clustering groups examples according to similarity. The resulting groups may help explore data, but are not automatically meaningful or useful.

How do training and test sets differ?

A training set is used to learn the model's patterns; a held-out test set estimates performance on new examples.

What can stratification preserve in a classification split?

A stratified split helps preserve the proportions of each class in both training and test sets.

What is data leakage?

Data leakage occurs when test-set information is used during training or preprocessing, making evaluation results overly optimistic.

How should preprocessing avoid test-set leakage?

Learn preprocessing transformations, such as filling missing values or scaling features, using training data only; then apply them to test data.

How should validation and test sets be used?

A validation set or cross-validation can help compare model choices during development. Reserve the test set for a final evaluation.

What does MAE measure in regression?

Mean absolute error (MAE) is the average absolute difference between predicted and actual values, expressed in the target's units.