🛒 Customer Segmentation - Online Retail

An unsupervised machine learning project for customer segmentation using K-Means and DBSCAN clustering algorithms.

🎯 Objective

Segment customers into meaningful groups based on their purchasing behavior to:

• Identify high-value customers
• Discover hidden patterns in customer data
• Enable targeted marketing strategies
• Improve customer retention

📊 Dataset

Online Retail dataset containing transactional data:

Feature	Description
InvoiceNo	Unique invoice number
StockCode	Product code
Description	Product description
Quantity	Quantity purchased
InvoiceDate	Date of transaction
UnitPrice	Price per unit
CustomerID	Unique customer identifier
Country	Customer's country

🔄 Project Pipeline

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│  Data Loading   │───►│  Preprocessing  │───►│  Feature Eng.   │
└─────────────────┘    └─────────────────┘    └─────────────────┘
                                                      │
                                                      ▼
┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│    Analysis     │◄───│   Clustering    │◄───│ Feature Scaling │
└─────────────────┘    └─────────────────┘    └─────────────────┘

🛠️ Techniques Used

Data Preprocessing

• Missing value handling
• Outlier detection and treatment (IQR method)
• Data type conversions
• Removing cancelled transactions

Feature Engineering (RFM Analysis)

• Recency: Days since last purchase
• Frequency: Number of transactions
• Monetary: Total spending

Clustering Algorithms

K-Means Clustering

• Optimal K selection using Elbow Method
• Silhouette Score analysis
• Centroid-based partitioning

DBSCAN Clustering

• Density-based spatial clustering
• Automatic outlier detection
• No need to specify number of clusters

Visualization

• Cluster distribution plots
• 2D/3D scatter plots
• Customer segment profiles

📁 Project Structure

online-retail-segmentation/
├── README.md
├── requirements.txt
├── notebooks/
│   └── customer_segmentation.ipynb
├── data/
│   └── .gitkeep (add your data here)
└── outputs/
    └── cluster_profiles.csv

🚀 Getting Started

Prerequisites

pip install -r requirements.txt

Required Libraries

pandas
numpy
scikit-learn
matplotlib
seaborn

Usage

1. Clone the repository

git clone https://github.com/YOUR_USERNAME/online-retail-segmentation.git
cd online-retail-segmentation

2. Add your dataset to data/ folder

3. Open and run the Jupyter notebook

jupyter notebook notebooks/customer_segmentation.ipynb

📈 Customer Segments

Example Segment Profiles

Segment	Recency	Frequency	Monetary	Description
Champions	Low	High	High	Best customers, buy often
Loyal	Low	Medium	Medium	Regular customers
At Risk	High	Medium	Medium	Haven't bought recently
Lost	High	Low	Low	Haven't bought in long time

🔍 Key Findings

1. Customer Distribution: Majority fall into 3-4 main segments
2. High-Value Customers: ~20% of customers drive ~80% of revenue
3. Seasonal Patterns: Purchasing behavior varies by season
4. Geographic Trends: Different countries show different patterns

💡 Business Applications

• Personalized Marketing: Target each segment differently
• Inventory Management: Stock based on segment preferences
• Customer Retention: Focus on at-risk segments
• Pricing Strategy: Offer deals to specific segments

📚 Learning Outcomes

This project demonstrates:

• Unsupervised learning techniques
• RFM (Recency, Frequency, Monetary) analysis
• Cluster validation methods
• Business insight generation from data

👨‍💻 Author

• IoT & AI Developer @ VoltX
• CS Student @ Helwan University

---

⭐ Star this repo if you find it useful!