/CustomerChurnPrediction_ANN

Artificial Neural Networks Models were created to predict whether a customer is a fraud or not.

Primary LanguagePythonMIT LicenseMIT

made-with-python

Customer Churn Prediction using ANN

Overview

Artificial Neural Networks Models were created to predict whether a customer is a fraud or not.

Imbalanced dataset handling methods were used for better accuracy.

Methods to handle imbalanced dataset

  • undersampling majority class
  • oversampling minority class
  • Synthetic Minority Oversampling Technique (uses KNN)

Models built

  • Dense Neural Networks model with 1 hidden layer
  • Dense Neural Networks model with 2 hidden layers

Prerequisites

  • Python 3
  • an IDE
  • Pandas
  • Numpy
  • Matplotlib
  • Sklearn
  • Tensorflow
  • Keras
  • Imblearn

All the external dependencies can be installed by using pip install -r requirements.txt

Author

  • Rishikesh Sivakumar

ForTheBadge built-with-love by Rishikesh Sivakumar