Performance Analysis of Algorithms for Credit Card Fraud Detection

In today's age we are seeing rapid integration of technology within our daily lives. The term “plastic money” has become a well-known entity which enables cashless transaction. Credit Cards plays a massive role in that entity. The number of credit card users is increasing day by day due to efficiency and card offers. Users are so much dependent on it either they want to buy a small home appliance of 5000 or a dream car of 1 crore. But in this generation where we are developing so fast in term of technology, the consequences or threats also come across with it. Credit Card fraud is one such fraud which has risen parallelly with the use of cashless technology. This project aims at detecting such credit card frauds. It uses four different machine learning algorithms to identify whether the transaction is genuine or fraud and makes comparison between the algorithms based on their functionality. The four algorithms are Ada-Boost, K-Nearest Neighbour, Random Forrest, and Logistic Regression and the accuracy of all algorithms was 99%.