Ai Based Credit Scoring System With Dynamic Risk Assessment

Credit cards are now potentially the most popularmode of payment for both offline and onlinepurchases thanks to new developments inelectronic commerce systems andcommunication technology; as a result, there ismuch more fraud involved with suchtransactions. Every year, fraudulent credit cardtransactions cause businesses and individuals tolose a lot of money, and con artists are constantlylooking for new tools and techniques to commitfraud. Researchers face a difficult task whentrying to identify credit card theft since criminalsare quick-thinking and inventive. The datasetprovided for credit card fraud detection isseverely unbalanced, making it difficult for thesystem to detect fraud. The use of credit cards isquite important in today's economy. It is anessential component of every family, company,and global enterprise. While using credit cardsresponsibly and safely can have many benefits,engaging in fraudulent behaviour can have anegative impact on your credit and finances.There have been several solutions proposed toaddress the escalating credit card theft. Theincreased use of electronic payments is nowsignificantly impacted by the detection offraudulent transactions. As a result, methods thatare efficient and effective for identifying fraud incredit card transactions are required. GradientBoosting Classifier, a machine learningmethodology, is suggested in this research as asmart method for identifying fraud in credit cardtransactions. The experimental results show thatthe suggested approach worked better than othermachine learning algorithms and reached themaximum accuracy performance, with trainingaccuracy of 100% and test accuracy of 91%