This repository contains a simple yet effective neural network built with PyTorch for customer segmentation based on demographic and behavioral features.

customer-segmentation-NN/
│
├── data/                        # Dataset location (after download)
│   └── external/
│       └── customer.zip
│
├── models/
│   ├── base_model.py           # CustomerSegmentationModel class
│   ├── neural_net.py           # PyTorch neural network architecture
│   └── init.py                 # Main script for training and evaluation
│
├── utils/
│   ├── get_data.py             # Download utility
│   └── zip.py                  # ZIP extraction utility
│
├── run.py                      # Entry point to automate the full pipeline
└── README.md                   # Project documentation

• Downloads customer segmentation dataset from Kaggle
• Preprocesses demographic and behavioral features
• Encodes categorical variables
• Trains a feedforward neural network with dropout and ReLU
• Evaluates performance (accuracy, classification report)
• Computes basic feature importance based on model weights

git clone https://github.com/your-username/customer-segmentation-NN.git
cd customer-segmentation-NN

It's recommended to use a virtual environment:

pip install -r requirements.txt

python src/init.py

This will:

• Download the dataset
• Extract it
• Train the model
• Display evaluation results and feature importance

• Input → Linear(128) → ReLU → Dropout
• → Linear(64) → ReLU → Dropout
• → Linear(output classes)

• Training Loss and Accuracy plots
• Test Accuracy
• Classification Report
• Feature Importance Ranking
• Sample Predictions with Probabilities

Dataset source: Kaggle - Customer Segmentation Dataset

Features include:

• Demographic: Gender, Age, Married status, Education, Profession, etc.
• Behavioral: Spending Score, Work Experience

This project is licensed under the MIT License. Feel free to use, modify, and contribute!