Saturday, 7 March 2026

📊 Day 47: Mosaic Plot in Python

Samaksh Dubey March 07, 2026 Data Science No comments

Day 47: Mosaic Plot in Python

On Day 47 of our Data Visualization series, we explored a powerful chart for analyzing relationships between categorical variables — the Mosaic Plot.

When you want to understand how two (or more) categorical variables interact with each other, Mosaic Plots provide a clear and intuitive visual representation.

Today, we applied it to the classic Iris dataset to examine the relationship between Species and Petal Size category.

🎯 What is a Mosaic Plot?

A Mosaic Plot is a graphical method for visualizing contingency tables (cross-tabulated categorical data).

It represents:

Categories as rectangles
Width proportional to one variable
Height proportional to another variable
Area representing frequency or proportion

👉 The larger the rectangle, the higher the frequency of that category combination.

📊 Dataset Used: Iris Dataset

The Iris dataset contains:

Sepal Length
Sepal Width
Petal Length
Petal Width
Species (Setosa, Versicolor, Virginica)

For this visualization, we:

Used Species as one categorical variable
Converted Petal Length into 3 categories:
- Small
- Medium
- Large

This helps us visually compare petal size distribution across species.

🧑‍💻 Python Implementation

✅ Step 1: Import Libraries


import pandas as pd
import matplotlib.pyplot as plt
from statsmodels.graphics.mosaicplot import mosaic
from sklearn.datasets import load_iris

Pandas → Data manipulation
Matplotlib → Plot rendering
Statsmodels → Mosaic plot function
Scikit-learn → Dataset loading

✅ Step 2: Load and Prepare Data


iris = load_iris()
df = pd.DataFrame(iris.data, columns=iris.feature_names)
df["Species"] = iris.target_names[iris.target]

We convert numeric species labels into readable category names.

✅ Step 3: Create Petal Size Categories


df["Petal Size"] = pd.cut(
    df["petal length (cm)"],
    3,
    labels=["Small", "Medium", "Large"]
)

Here we divide petal length into three equal bins.

This transforms a continuous variable into a categorical one.

✅ Step 4: Create the Mosaic Plot


plt.figure(figsize=(9, 5))
mosaic(df, ["Species", "Petal Size"], gap=0.02)
plt.title("Mosaic Plot: Species vs Petal Size")
plt.tight_layout()
plt.show()

Key Parameters:

["Species", "Petal Size"] → Defines categorical relationship
gap=0.02 → Adds spacing between tiles
figsize → Controls plot size

📈 What the Visualization Reveals

From the Mosaic Plot:

🌸 Setosa

Almost entirely in the Small petal category
Very little variation

🌿 Versicolor

Mostly in the Medium category
Some overlap into Small and Large

🌺 Virginica

Dominantly in the Large category
Some presence in Medium

🔍 Key Insight

The Mosaic Plot clearly shows that:

Petal size is strongly associated with species
Species are well separated based on petal length
This confirms why petal measurements are highly important features in classification models

Even without machine learning, we can visually detect separation patterns.

💡 Why Use Mosaic Plots?

✔ Excellent for categorical comparisons
✔ Shows proportional relationships clearly
✔ Works well with contingency tables
✔ Helpful in statistical analysis
✔ Easy to interpret once understood