Project Overview

In this hands-on project, we'll build a Customer Segmentation Model using K-Means Clustering. The goal is to group customers based on their annual income and spending behavior, helping businesses make smart, data-driven decisions.

Project Objectives

• Understand customer data and clean it.
• Perform Exploratory Data Analysis (EDA).
• Apply scaling and preprocessing.
• Use the Elbow Method to choose optimal K.
• Apply K-Means Clustering.
• Evaluate cluster quality using Silhouette Score.
• Visualize and interpret customer segments.

Prerequisites

pip install pandas numpy matplotlib seaborn scikit-learn

Project Workflow

Step 1: Import Libraries

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

from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import silhouette_score

Step 2: Create / Load Dataset

For demonstration, we'll create a sample dataset of mall customers.

data = {
    "CustomerID": list(range(1, 21)),
    "Age":        [19, 21, 20, 23, 31, 22, 35, 45, 60, 55,
                   30, 40, 28, 33, 50, 27, 42, 38, 47, 25],
    "Annual_Income": [15, 16, 17, 18, 20, 25, 35, 40, 50, 55,
                      45, 60, 70, 75, 80, 90, 95, 100, 110, 120],   # in $1000
    "Spending_Score": [39, 81, 6, 77, 40, 76, 6, 94, 3, 72,
                       55, 65, 50, 60, 20, 90, 30, 85, 25, 95]   # 1-100
}

df = pd.DataFrame(data)
print(df.head())

Step 3: Data Exploration

print(df.shape)
print(df.info())
print(df.describe())
print(df.isnull().sum())

Visualize the distribution of each feature:

sns.pairplot(df[["Age", "Annual_Income", "Spending_Score"]])
plt.show()

Step 4: Feature Selection

For segmentation, we'll use two key features:

X = df[["Annual_Income", "Spending_Score"]]
print(X.head())

Step 5: Feature Scaling

scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)

Step 6: Find Optimal K Using Elbow Method

wcss = []
for k in range(1, 11):
    km = KMeans(n_clusters=k, init="k-means++", random_state=42, n_init=10)
    km.fit(X_scaled)
    wcss.append(km.inertia_)

plt.plot(range(1, 11), wcss, marker="o", color="red")
plt.title("Elbow Method")
plt.xlabel("Number of Clusters (K)")
plt.ylabel("WCSS")
plt.grid(True)
plt.show()

Step 7: Apply K-Means with Optimal K

kmeans = KMeans(n_clusters=5, init="k-means++", random_state=42, n_init=10)
y_kmeans = kmeans.fit_predict(X_scaled)

df["Cluster"] = y_kmeans
print(df.head())

Step 8: Evaluate Cluster Quality

score = silhouette_score(X_scaled, y_kmeans)
print(f"Silhouette Score: {score:.3f}")

Step 9: Visualize Clusters

plt.figure(figsize=(8, 6))
sns.scatterplot(
    x=X["Annual_Income"], y=X["Spending_Score"],
    hue=y_kmeans, palette="viridis", s=100
)
plt.scatter(
    scaler.inverse_transform(kmeans.cluster_centers_)[:, 0],
    scaler.inverse_transform(kmeans.cluster_centers_)[:, 1],
    s=200, c="red", marker="X", label="Centroids"
)
plt.title("Customer Segments")
plt.xlabel("Annual Income (k$)")
plt.ylabel("Spending Score")
plt.legend()
plt.show()

Step 10: Interpret Customer Segments

Based on income and spending behavior, here are typical clusters:

Business Insights & Strategies

Bonus: Save the Model

import joblib

joblib.dump(kmeans, "customer_segmentation_model.pkl")
joblib.dump(scaler, "scaler.pkl")

# Load and Predict
loaded_model = joblib.load("customer_segmentation_model.pkl")
loaded_scaler = joblib.load("scaler.pkl")

new_customer = [[70, 80]]  # Annual Income, Spending Score
prediction = loaded_model.predict(loaded_scaler.transform(new_customer))
print("Predicted Cluster:", prediction[0])

Suggested Project Folder Structure

customer_segmentation_project/
│
├── data/
│   └── customers.csv
├── notebooks/
│   └── customer_segmentation.ipynb
├── models/
│   └── customer_segmentation_model.pkl
├── visualizations/
│   ├── elbow_plot.png
│   └── cluster_plot.png
├── requirements.txt
└── README.md

Real-Life Analogy

Real-World Applications

Common Mistakes to Avoid

Best Practices

What You Learned

Golden Rule