Scribd
Upload
22 views3 pages

PCA 4D To 2D Full Solution

The document details a step-by-step application of Principal Component Analysis (PCA) to reduce a 4-dimensional dataset to 2 dimensions. It includes standardization of the data, calculation of the covariance matrix, extraction of eigenvalues and eigenvectors, selection of top eigenvectors, and projection of the data onto these principal components. The final output retains 96.5% of the variance in the reduced 2D data.

Uploaded by

waseem
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
22 views3 pages

PCA 4D To 2D Full Solution

The document details a step-by-step application of Principal Component Analysis (PCA) to reduce a 4-dimensional dataset to 2 dimensions. It includes standardization of the data, calculation of the covariance matrix, extraction of eigenvalues and eigenvectors, selection of top eigenvectors, and projection of the data onto these principal components. The final output retains 96.5% of the variance in the reduced 2D data.

Uploaded by

waseem
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

PCA Numerical Example (4D to 2D)

PCA Numerical Example (4D to 2D) — Fully Detailed Step-by-Step Solution

GOAL

We apply Principal Component Analysis (PCA) to reduce 4-dimensional data to 2


dimensions, with complete detailed steps.

Given Data Matrix X — 5 Samples, 4 Features

X=[

[4, 2, 0, 1],

[2, 4, 0, 3],

[2, 2, 2, 3],

[4, 4, 2, 5],

[6, 6, 4, 7]

Step 1: Standardize the Data (Z-score Normalization)

Mean of features (μ):

μ = [3.6, 3.6, 1.6, 3.8]

Standard deviation (σ):

σ ≈ [1.673, 1.673, 1.673, 2.280]

Standardized matrix Z ≈

[
[0.239, -0.956, -0.956, -1.228],

[-0.956, 0.239, -0.956, -0.351],

[-0.956, -0.956, 0.239, -0.351],

[0.239, 0.239, 0.239, 0.526],

[1.433, 1.433, 1.433, 1.404]

Step 2: Covariance Matrix (C)

C≈

[1, 0.83, 0.83, 0.78],

[0.83, 1, 0.83, 0.78],

[0.83, 0.83, 1, 0.78],

[0.78, 0.78, 0.78, 1]

Step 3: Eigenvalues and Eigenvectors

Eigenvalues ≈ [3.37, 0.49, 0.13, 0.01]

Total variance = 4

Explained Variance:

PC1 ≈ 84.2%

PC2 ≈ 12.3%

Together: 96.5%

Step 4: Select Top 2 Eigenvectors (Principal Components)


W≈

[0.51, -0.62],

[0.51, -0.62],

[0.51, 0.46],

[0.47, 0.09]

Step 5: Project Data onto Principal Components

Z_PCA ≈

[-1.49, -0.34],

[-1.04, 0.61],

[-1.09, -0.45],

[0.42, 0.41],

[3.20, -0.23]

Final Output: 2D data with 96.5% variance retained.

You might also like