Machine Learning Algorithm Evaluation for Detection of Fake Bank Currency

Abstract

The presence of counterfeit banknotes in the financial market poses a significant challenge to the integrity of a country's currency system. Differentiating between genuine and forged notes manually is a daunting task due to the high precision employed by miscreants in creating fake currency. To address this issue, arandom system is proposed for banks and ATM machines which predicts the authenticity of banknotes. In this research paper, we will explore the application of Supervised Machine Learning Algorithm.Dataset was sourced from the UCI machine learning system. Additionally, we introduce the Light-GBM algorithm and analysed its performance against the other algorithms.