Credit Card Fraud Detection using Machine Learning
V.Sellam1, P.Tushar2, G.Rohit3, S.Sanyam4

1V.Sellam, Asst Professor, Department of CSE, SRM University, Ramapuram, Chennai.
2P.Tushar, Student, Computer Science Engineer, SRM University, Ramapuram, Chennai.
3G.Rohit, Student, Computer Science Engineer, SRM University, Ramapuram, Chennai,
4S.Sanyam, Student, Computer Science Engineer, SRM University, Ramapuram, Chennai.
Manuscript received on January 1, 2021. | Revised Manuscript received on January 09, 2021. | Manuscript published on January 10, 2021. | PP: 16-19 | Volume-1 Issue-1, February 2021 | Retrieval Number:A1003011121/2021©LSP
Open Access | Ethics and Policies
© The Authors. Published by Lattice Science Publication (LSP). This is an open access article under the CC BY-NC-ND license (

Abstract: These days online transactions are the most Preferred mode of transactions. It’s basically a constant payment method which has become a key part of our lives. But there are some problems associated with this mode of transaction which are fraud transactions that are associated with it and as the count of the online transactions increase, the count of the fraudulent transactions increases along with it. If not being completely put to an end, these. Fraud transactions can definitely be reduced to some extent. There are various methods for that, out of which data analytics and machine learning are one of the methods First a data set is provided from which the maximum, minimum and standard deviation is found. Using this a histogram is plotted which provides a visualisation of the data. Once this is done, 2 groups of graphs are created using the data which are the amount vs class graph and type of transactions vs time graph. Then later 3 machine learning algorithms are used that is light GBM , Adaboost and random forrest classifier to provide the recall , precision and accuracy of the model. A function to find the time taken to run these algorithms is also used. In the end, the value provided by these 3 algorithms are compared to find the one which provides the best result.
Keywords: Fraud transactions, GBM , Adaboost and random forrest classifier.