Fraud Detection in Financial Domain using Machine Learning

Abstract

Digital fraud has become a menace in every industry. It is critical for any firm to have a concentrated focus on detecting and preventing fraudulent activities. Security is a priority. The way we communicate has changed dramatically as a result of digitization. A simple click of a mouse, complete our day-to-day transactions. On the other hand, it has created concerns from swindlers who take advantage of absent protections in current financial systems and mimic real customers, undertake time-consuming transactions on their behalf that result in a profit causing financial setbacks to the organizations and customers. Organizations will need to pay attention as a result of this. Its brand value is also affected. Organizations have learned from their mistakes. To prevent fraud and keep ahead of the criminals, it is necessary to maintain a constant focus. It’s critical to keep an eye on major trends. We might be able to tell the difference between a legitimate and a fraudulent transaction, obtaining customer data such as geolocation, authentication, and so on, it is possible to keep track of the device’s IP address during the session. Machine Learning (ML) will assume a significant part in the future in identifying examples of such frauds consequently. We use algorithms like decision trees, XGBoost, K-NN and others to find an optimal solution for our concerning project.