📊 Day 17: Pair Plot (Scatter Matrix) in Python

 

📊 Day 17: Pair Plot (Scatter Matrix) in Python

🔹 What is a Pair Plot?

A Pair Plot (also called a Scatter Matrix) displays pairwise relationships between multiple numerical variables in a dataset.
It combines scatter plots and distribution plots into a single grid.

---

🔹 When Should You Use It?

Use a pair plot when:

• Performing exploratory data analysis (EDA)

• Studying relationships between multiple variables

• Identifying correlations, clusters, and trends

• Detecting outliers

---

🔹 Example Scenario

Suppose you are analyzing:

• Iris dataset

• Customer behavior metrics

• Financial indicators

A pair plot helps you instantly see:

• Relationships between every feature

• Feature distributions

• Possible feature interactions

---

🔹 Key Idea Behind It

👉 Each cell shows a relationship between two variables
👉 Diagonal shows distribution of individual variables
👉 Off-diagonal cells show scatter plots

---

🔹 Python Code (Pair Plot)

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np 

data = np.random.rand(100, 4)
df = pd.DataFrame(data, columns=['A', 'B', 'C', 'D'])

sns.pairplot(df)

plt.show()

---

🔹 Output Explanation

• Diagonal plots show histograms or density plots

• Off-diagonal plots show scatter relationships

• Patterns reveal correlations or independence

• Outliers are easily noticeable

---

🔹 Pair Plot vs Correlation Heatmap

Feature	Pair Plot	Correlation Heatmap
Visual detail	High	Medium
Exact values	No	Yes
Relationship view	Scatter-based	Color-based
Best for	Deep EDA	Quick overview

---

🔹 Key Takeaways

• Pair plots give a complete relationship overview

• Best used in early data exploration

• Powerful for feature understanding

• Avoid using with too many variables