📊 ANN Customer Churn Prediction

🚀 Live Application

👉 Deployment Link

🚀 Live Application

👨‍💻 Author

Name	Education	Specialization
Ravi Raj	B.Tech 3rd Year	Artificial Intelligence & Machine Learning

---

📌 Project Overview

Customer churn prediction helps businesses identify customers who are likely to leave their services.

This project uses an Artificial Neural Network (ANN) model built with TensorFlow/Keras to predict whether a bank customer will churn or not.

The model is deployed using Streamlit, allowing users to interactively input customer details and get predictions instantly.

---

🎯 Objective

The goal of this project is to:

• Predict whether a customer will leave the bank
• Help businesses reduce customer churn
• Build a real-world Deep Learning application
• Deploy the model as an interactive web app

---

🧠 Machine Learning Workflow

Steps	Description
Data Collection	Customer churn dataset
Data Preprocessing	Encoding categorical features
Feature Scaling	StandardScaler used
Model Building	Artificial Neural Network
Model Training	TensorFlow / Keras
Model Deployment	Streamlit Web App

---

# 🏗️ Tech Stack

|      Technology    |        Usage        |
|--------------------|---------------------|
| Python             | Programming         |
| TensorFlow / Keras | Deep Learning       |
| Pandas             | Data processing     |
| NumPy              | Numerical operations|
| Scikit-Learn       | Preprocessing       |
| Streamlit          | Web App             |
| GitHub             | Version Control     |

---

📊 Features Used

Feature	Description
CreditScore	Customer credit score
Geography	Customer location
Gender	Male / Female
Age	Customer age
Tenure	Years with bank
Balance	Account balance
NumOfProd.	Number of bank products
HasCrCard	Credit card ownership
IsAct.Member	Active membership
EstimatedSalary	Customer estimated salary

---

🧩 Model Architecture

Layer	Details
Input Layer	Customer Features
Hidden Layer 1	Dense + ReLU
Hidden Layer 2	Dense + ReLU
Output Layer	Sigmoid (Binary Classification)

---

# 📂 Project Structure
ANN-churn-prediction
│
├── app.py
├── eda.ipynb
├── model.h5
├── scaler.pkl
├── label_encoder_gender.pkl
├── onehot_encoder_geo.pkl
├── requirements.txt
├── Churn_Modelling.csv
└── README.md


---

⚙️ Installation

Clone the repository

git clone https://github.com/ravicoder01/ANN-churn-prediction.git

Go to project directory

cd ANN-churn-prediction

Install dependencies

pip install -r requirements.txt

Run the Streamlit application

streamlit run app.py

---

📈 Example Prediction

Input	Example
Credit Score	650
Geography	France
Gender	Male
Age	35
Balance	60000

Output

Prediction: Customer likely to stay

or

Prediction: Customer likely to churn

---

🌍 Real World Applications

Industry	Use Case
Banking	Predict customer account closure
Telecom	Predict user switching
SaaS	Subscription churn detection
E-commerce	Identify inactive customers

---

📚 Learning Outcomes

Through this project I learned:

• Artificial Neural Networks
• Data preprocessing techniques
• Model serialization
• Machine Learning deployment
• Building interactive apps with Streamlit

---

⭐ Support

If you found this project useful, please give it a star ⭐ on GitHub.

Repository Link:

https://github.com/ravicoder01/ANN-churn-prediction

---