The purpose of this project was to detect if fraud had been committed on a credit card using a sample taken over two days. EDA was conducted on the dataset to detect visually if any trends existed. Resampling of the dataset was also conducted to remedy its stark imbalance. Various traditional machine learning algorithms were then evaluated, along with deep neural networks, to decide on the most efficient and accurate model to predict fraud. It will be very helpful to whoever finds this interesting to take a look at the full report available at the Project website url above under 'Artificial Intelligence Capstone'.