Customer Churn Prediction Dashboard

A comprehensive Machine Learning project that predicts customer churn using multiple ML algorithms, with a beautiful Node.js/React dashboard for visualization and real-time predictions.

Features

Machine Learning

• 9+ ML Algorithms: Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, XGBoost, LightGBM, SVM, KNN, Naive Bayes, Neural Network
• Comprehensive Metrics: Accuracy, Precision, Recall, F1-Score, ROC-AUC
• Feature Importance Analysis
• Customer Segmentation Insights

Dashboard

• Interactive Visualizations: Charts, graphs, and tables
• Model Comparison: Side-by-side performance metrics
• ROC Curves: Visual model evaluation
• Confusion Matrices: Detailed prediction analysis
• Live Predictions: Predict churn for new customers
• Customer Insights: Segmentation by subscription, contract, age, gender
• Responsive Design: Works seamlessly across all devices

Project Structure

Customer_Churn_Prediction/
├── Dataset/
│   ├── customer_churn_dataset-training-master.csv
│   └── customer_churn_dataset-testing-master.csv
├── ml_model/
│   ├── train_model.py          # ML training pipeline
│   ├── predict.py              # Prediction service
│   ├── requirements.txt        # Python dependencies
│   ├── models/                 # Saved ML models
│   └── results/                # Generated metrics JSON
├── backend/
│   ├── server.js               # Express API server
│   └── package.json
├── frontend/
│   ├── src/
│   │   ├── App.js              # Main React component
│   │   ├── index.js
│   │   └── index.css           # Styling
│   ├── public/
│   └── package.json
├── templates/                  # Flask HTML templates
├── static/                     # Static assets
├── app.py                      # Flask application (alternative dashboard)
├── start_servers.ps1           # PowerShell startup script
├── start-servers.bat           # Batch startup script
└── README.md

Setup Instructions

Prerequisites

• Python 3.8+
• Node.js 16+ (for React dashboard)
• npm or yarn

Step 1: Install Python Dependencies

cd ml_model
pip install -r requirements.txt

Step 2: Train ML Models

cd ml_model
python train_model.py

This will:

• Train 9+ ML models
• Generate performance metrics
• Save models to ml_model/models/
• Save dashboard data to ml_model/results/dashboard_data.json

Option A: Node.js/React Dashboard (Recommended)

Step 3A: Start Backend Server

cd backend
npm install
npm start

The API server will run on http://localhost:5000

Step 4A: Start Frontend Dashboard

cd frontend
npm install
npm start

The dashboard will open at http://localhost:3000

Quick Start (PowerShell)

.\start_servers.ps1

Quick Start (Command Prompt)

start-servers.bat

Option B: Flask Dashboard

pip install flask flask-cors
python app.py

The Flask dashboard will run on http://localhost:5000 with integrated UI at /dashboard

Screenshots

Dashboard Overview

The main dashboard displays key metrics and visualizations for churn analysis.

Analytics Dashboard

Overview tab with comprehensive visualizations including model accuracy comparison, churn distribution, feature importance, and age-based churn analysis.

Model Performance Ranking

Detailed ranking of all 11 ML models with accuracy, precision, recall, F1-score, and ROC-AUC metrics.

Customer Insights

Segmentation analysis showing churn patterns by subscription type, contract length, and risk factors.

API Endpoints

Endpoint	Method	Description
/api/health	GET	Health check
/api/dashboard	GET	All dashboard data
/api/stats	GET	Dataset statistics
/api/models	GET	Model results & ranking
/api/roc-curves	GET	ROC curve data
/api/confusion-matrices	GET	Confusion matrices
/api/feature-importance	GET	Feature importance
/api/segments	GET	Customer segments
/api/churn-factors	GET	Churn risk factors
/api/predict	POST	Predict customer churn

Dashboard Features

Overview Tab

• Dataset statistics cards
• Model accuracy comparison chart
• Churn distribution pie chart
• Feature importance ranking
• Age group churn analysis

Model Comparison Tab

• Full model ranking table
• Multi-metric bar charts
• ROC curves overlay
• Confusion matrices for top models

Insights Tab

• Churn by subscription type
• Churn by contract length
• Churn risk factors analysis
• Gender-based segmentation

Predict Churn Tab

• Interactive prediction form
• Model selection dropdown
• Real-time churn prediction
• Risk level classification

Tech Stack

• ML Pipeline: Python, Scikit-learn, XGBoost, LightGBM
• Backend: Node.js, Express.js
• Frontend: React, Recharts
• Styling: Custom CSS

Dataset Features

Feature	Type	Description
CustomerID	Numeric	Unique identifier
Age	Numeric	Customer age
Gender	Categorical	Male/Female
Tenure	Numeric	Months with company
Usage Frequency	Numeric	Service usage frequency
Support Calls	Numeric	Number of support calls
Payment Delay	Numeric	Payment delay in days
Subscription Type	Categorical	Basic/Standard/Premium
Contract Length	Categorical	Monthly/Quarterly/Annual
Total Spend	Numeric	Total amount spent
Last Interaction	Numeric	Days since last interaction
Churn	Target	0 = Stayed, 1 = Churned

Model Performance

Models are automatically ranked by accuracy. Typical results include:

• Best models usually achieve 85-90%+ accuracy
• Feature importance helps identify key churn predictors
• ROC-AUC scores indicate model discrimination ability

License

This project is for educational purposes.