Customer Segmentation using K-Means Clustering

This project implements a customer segmentation analysis using K-Means clustering. It groups customers based on their behavioral and financial attributes (Age, Annual Income, and Spending Score) to help identify distinct market segments.

Features

• Synthetic Data Generation: Automatically generates a realistic dataset if none exists.
• Data Preprocessing: Handles feature scaling using StandardScaler.
• Optimal Cluster Determination:
  • Elbow Method: Visualization to determine the optimal $K$.
  • Silhouette Score: Quantitative metric to validate cluster separation.
• K-Means Clustering: Groups customers into optimized segments.
• Visualization:
  • 2D Scatter Plots (Income vs. Spending Score, etc.)
  • 3D Scatter Plots (Age vs. Income vs. Spending Score)
• Insight Extraction: Exports segmented data to CSV and prints cluster profiles.

Project Structure

Customer_Segmentation/
├── data/
│   ├── customer_data.csv       # Input dataset (generated)
│   └── segmented_customers.csv # Output with Cluster labels
├── src/
│   ├── data_loader.py          # Data generation and loading
│   ├── preprocessing.py        # Feature scaling
│   ├── clustering.py           # K-Means logic
│   └── visualization.py        # Plotting functions
├── main.py                     # Entry point script
├── requirements.txt            # Python dependencies
└── README.md                   # Project documentation

Installation

1. Clone the repository (if applicable) or navigate to the project folder.

2. Install dependencies:

   pip install -r requirements.txt

Usage

Run the main analysis script:

python main.py

What to Expect

1. The script will generate data/customer_data.csv if it doesn't exist.
2. It will display the Elbow Method and Silhouette Score plots.
   • Note: You must close these plot windows for the script to continue.
3. It will automatically select the optimal number of clusters (typically between 3 and 6).
4. It will perform clustering and display 2D and 3D visualizations.
5. Finally, it saves the results to data/segmented_customers.csv and prints a summary of each cluster's characteristics.

Technologies Used

• Python 3.x
• Pandas (Data manipulation)
• Scikit-learn (Clustering and Preprocessing)
• Matplotlib & Seaborn (Visualization)
• NumPy (Numerical operations)