Easy Leave Application

Abstract

The objective of this research is to provide a regression algorithm-based machine learning anomaly solution for banks to anticipate credit card fraud. This project intends to create a web application that analyses patterns in transaction data to forecast credit card fraud. the system uses regression techniques, such as linear regression and logistic regression, as well as abnormality detection algorithms, such as anomaly detection, to detect probable fraud. the analysis' findings are then given to bank staff in an intuitive online application so they may review them and take appropriate action. the findings demonstrate that the suggested method helps banks lessen their losses from fraudulent transactions and provide precise fraud forecasts. this study shows how machine learning algorithms may be used to detect and prevent credit card fraud and can be a useful tool for banks to enhance their fraud management procedures. the characteristics used to determine whether or not a transaction is fraudulent include the old and new account balances as well as additional fields. aws (amazon web services) ses (simple email service) cloud will notify the bank if the transaction is fraudulent. the current system was not developed using real-time transaction datasets, and it also has low accuracy and low efficiency in terms of loading time and implementation time. when compared to the current system, the suggested system's loading and execution speeds are very quick. the suggested method may be further enhanced for complicated use cases and is very effective and scalable.