Overview

The repository contains an Artificial Neural Network (ANN)-based Deep Learning model for predicting the churn classifier of a Financial Institution, i.e. Bank. A churn predictor for a bank is developed using Artificial Neural Network architecture.

What is Churn Rate

Churn rate measures the number of customers or employees who leave the bank during a given period.

Necessary Tools

• TensorFlow
• Keras
• Numpy
• Pandas
• Matplotlib

Optimizer Used

• Adam

Activation Functions Used

• ReLU
• Sigmoid

Neural Network Architechture

IL: Neuron 11 HL1: Neuron 7 HL2: Neuron 6 OL: Neuron 1 Dropout: 0.1

Training

Early Stopping

Testing and Validation

• Accuracy: 85.85%