An Explainable Machine Learning Model to Analyze and Detect Fake Currency

The advent of state-of-the-art techniques and technologies has exposed human beings to some threats as well, apart from providing several advantages. Fake currency is such a major threat to society which in turn obstructs the country's economic growth and financial status and causes an impact on human life. In addition to it, detecting fake currency is very challenging for the human eye. Detection of such currency is an important task and also consumes a lot of time if done manually. To address this critical issue, our paper contributes to identifying fake currency using Machine learning (ML) algorithms. In our article, we experimented with multiple ML algorithms to detect fake currencies and analyze their performance. Among the six algorithms experimented with for our work, Random Forest has shown the highest accuracy with 99.27% and a precision of 0.99 compared to the other algorithms. Additionally, we have demonstrated the interpretability of the ML model using LIME and SHAP which would be beneficial in understanding the important factors behind the particular prediction. We also performed hyperparameter tuning using GridSearchCV and manually too which showed us an improved accuracy performance of 99.99% from 99.27% for Random Forest. Feature Importance analysis and Explainability concepts are also demonstrated to provide an in-depth analysis of model prediction.