This project presents an ANN-based Customer Churn Prediction System designed to identify customers who are likely to leave a service, while also incorporating a salary regression model to support deeper customer analytics. The primary objective of the project is to leverage Artificial Neural Networks (ANN) to build an intelligent, data-driven solution that aids organizations in improving customer retention and decision-making.
Customer churn is a critical business problem, as acquiring new customers is often more expensive than retaining existing ones. To address this challenge, an ANN classifier was developed to predict whether a customer will churn or remain active based on various demographic and behavioral features. The classification model uses multiple fully connected layers with ReLU activation functions in the hidden layers and a Sigmoid activation function in the output layer to generate churn probabilities. The model is trained using the Binary Crossentropy loss function and optimized with the Adam optimizer, ensuring stable and efficient learning.
In addition to churn classification, the project includes an ANN-based salary regression model that predicts customer salary values using the same feature space. This regression model employs ReLU activation functions in the hidden layers and a Linear activation function in the output layer, making it suitable for continuous value prediction. The model is trained using Mean Squared Error (MSE) as the loss function and evaluated using metrics such as RMSE, MAE, and R² score. Although secondary to churn prediction, the salary regressor enhances the analytical depth of the system.
A complete data preprocessing pipeline was implemented, including handling missing values, encoding categorical variables, and feature scaling, which is essential for effective ANN performance. To make the solution accessible and interactive, the entire system was deployed using Streamlit, allowing users to input customer data and receive real-time churn and salary predictions through a web interface.
Overall, this project demonstrates an end-to-end machine learning workflow, combining data preprocessing, ANN modeling for both classification and regression, model evaluation, and web deployment. It highlights the practical application of Artificial Neural Networks in solving real-world business problems related to customer retention and predictive analytics.