This is to create feasbility models to detect frauds in credit card transactions using Deep Neural Network, Decision Tree and Random Forest. Sampling techniques are implemented to tackle class imbalance problem.
Dataset can be downloaded from Kaggle: https://www.kaggle.com/mlg-ulb/creditcardfraud
Customer clustering based on K-means and K-means + Autoencoder models are also available.
Production codes are updated for RF. ENJOY 💕💕
Dockerfile and Flask API dev are to be added. Stay tuned🤞🤞