Ch.03 Mini Project 03 — Correlation & Heatmap Analysis¶

Concept: Scatter plot matrix (SPLOM), correlation matrix visualization, correlogram, Pearson and Spearman correlation, heatmap with clustering.

What it does¶

Uses the Friends dataset (built-in) to demonstrate three correlation/heatmap visualization types:

Scatter plot matrix (SPLOM) — all pairs of numeric attributes shown as a grid of scatter plots; diagonal shows attribute name
Correlation heatmap — color-coded Pearson r matrix for all attribute pairs at once; values annotated inside cells
ASCII correlation table — text-only Pearson r table with significance stars (*** p<0.001, ** p<0.01, * p<0.05)
Spearman correlation matrix — rank-based alternative; also printed as ASCII and as a color heatmap

Answers the Iris dataset exercise questions Q10, Q11, Q12 from Ch.03, and also computes covariance/correlation for the Friends dataset:

Exercise	Plot type	R function
Q10	Scatter plot matrix with Pearson r values	GGally::ggpairs
Q11	Correlogram	corrplot::corrplot
Q12	Heatmap with clustering dendrogram	pheatmap::pheatmap
—	Friends covariance + correlation	base R cov(), cor()

File	Language	Description
`correlation_analysis.py`	Python 3	Pure stdlib + optional matplotlib
`correlation_analysis.R`	R	GGally, corrplot, pheatmap

# Built-in demo (Friends dataset)
python correlation_analysis.py

# From a CSV file
python correlation_analysis.py data.csv

No external dependencies for text output. Install matplotlib for plots:

pip install matplotlib

# Install required packages once:
install.packages(c("ggplot2", "GGally", "corrplot", "pheatmap"))

source("correlation_analysis.R")

Concept	Where it appears
Scatter plot matrix (SPLOM)	`plot_scatter_matrix()`
Pearson correlation heatmap	`plot_correlation_heatmap()`
Spearman correlation	`compute_spearman_matrix()`
Correlogram	R: corrplot, Python: heatmap with color scale
Heatmap + dendrogram	R: pheatmap

Pearson r:   r = cov(X,Y) / (std(X) * std(Y))   [linear only]
Spearman rho: apply Pearson r to the RANKS of X and Y  [monotonic, robust]