Customer Churn Prediction

Predicting customer churn using Deep Learning (Aritificial Neural Network)

Project Description 

Developed a Customer Churn Prediction System using an Artificial Neural Network (ANN) to identify customers who are likely to discontinue services.

The model was built and trained using TensorFlow, with data preprocessing and analysis performed using Pandas, NumPy, Scikit-learn, and Matplotlib.

The backend is implemented using Python Flask, where the trained ANN model is deployed and exposed via REST APIs. The frontend is developed using HTML, CSS, and JavaScript, enabling users to input customer details and receive real-time churn predictions.

Features 

• Predicts the probability of customer attrition.

• Deep learning model built using TensorFlow.

• Feature scaling, encoding, and normalization using Scikit-learn.

• Exploratory data analysis and performance metrics visualization using Matplotlib.

• Seamless frontend-backend communication via HTTP requests.

• User-friendly UI for inputting customer data and viewing predictions.

• Instant churn probability output after user input submission.

Tech Stack

• Python 3.x

• Flask

• TensorFlow / Keras

• Pandas

• NumPy

• Scikit-learn

• Matplotlib

• HTML5

• CSS3

• JavaScript

System Requirements

• Python 3.8+

• Anaconda

• GPU not required

Deliverables 

• Project Code

• Dataset

• Jupyter Notebook

• Setup over video call

• WhatsApp support