lesson-principal-component-analysis

Principal Component Analysis

---

Materials We Provide

Topic	Description	Link
Lesson	Principal Component Analysis Jupyter Notebook	Link
Slide Deck	Principal Component Analysis Slide Deck	Link

Dataset description: Wine Quality Dataset

---

Learning Objectives

After this lesson, students will be able to:

1. Differentiate between feature selection and feature extraction.
2. Describe the PCA algorithm.
3. Implement PCA in scikit-learn.
4. Calculate and interpret proportion of explained variance.
5. Identify use cases for PCA.

---

Student Requirements

Before this lesson(s), students should already be able to:

1. Construct a multiple linear regression model.
2. Interpret variance.
3. Know what a covariance or correlation matrix is.

---

Lesson Outline

Total Time: 120 minutes

I. PCA (120 minutes)

---

OPTIONAL: Resources for Practice and Learning

• Setosa.io PCA Visualization
• Medium Article: A One-Stop Shop for Principal Component Analysis
• Exceptional StackOverflow Answer to Understanding PCA
• A Semi-Academic Walkthrough of PCA Building Blocks
• Eigenvalues and Eigenvectors Video
• PCA Documentation in sklearn
• PCA Explanation on AnalyticsVidhya, includes R and Python code
• An Introduction to Statistical Learning with Applications in R, PCA is in Sections 6.3, 6.7, 10.2
• Notes on PCA from Penn State University’s Applied Multivariate Statistical Analysis Course
• Intuitive Description of Positive Definite Matrices
• What are Eigenvaalues and Eigenvectors? A Medium post.

• StackExchange: What is the problem with p < n?