Ch.03 Mini Project 02 — Multivariate Visualization

Concept: Parallel coordinates, star/radar plots, bubble charts, 3D scatter plots, heatmaps — all for visualizing multivariate data.

What it does

Python (multivariate_viz.py)

Uses the Friends dataset (built-in) to demonstrate five multivariate plot types:

1. Parallel coordinates — each friend is a polyline crossing vertical axes, colored by Company (Good/Bad)
2. Star plots — one radar/spider chart per friend showing their numeric profile
3. Bubble chart — Weight vs Height, bubble size = Max_temp, color = Gender
4. 3D scatter — Max_temp, Weight, Height in a 3D point cloud
5. Heatmap — color-coded grid showing all attribute values per friend

matplotlib is required for plots. If not available, a text description is printed.

R (multivariate_viz.R)

Answers the Iris dataset exercise questions from Ch.03:

Exercise	Plot type	R function
Q1	Scatter: Sepal.Length vs Sepal.Width; Petal.Length = size	ggplot2
Q2	Scatter: same axes; Species = color + shape	ggplot2
Q3	3D scatter: Sepal.Length, Sepal.Width, Petal.Length	scatterplot3d
Q4	Parallel coordinates; Species = color	MASS::parcoord
Q5	Star plots for first 20 objects	stars()
Q8	Chernoff faces for first 20 objects	aplpack::faces
Q9	Box plots for all 4 numeric attributes	ggplot2

Files

File	Language	Description
multivariate_viz.py	Python 3	matplotlib; beginner-friendly
multivariate_viz.R	R	ggplot2, scatterplot3d, MASS, aplpack

Usage

Python

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

Install matplotlib if needed:

pip install matplotlib

R

# Install required packages once:
install.packages(c("ggplot2", "scatterplot3d", "MASS", "aplpack"))

source("multivariate_viz.R")

Key Concepts from Ch.03 Applied

Concept	Function
Parallel coordinates	plot_parallel_coordinates()
Star / radar plots	plot_star_plots()
Bubble chart (4 attributes)	plot_bubble_chart()
3D scatter	plot_3d_scatter()
Heatmap	plot_heatmap()
Color / shape encoding	Color by class in all plots