Credit Card Fraud Detection Using Machine Learning

Credit Card Fraud Detection Using Machine Learning


  • Ms. Merin Manoj ktu
  • Akshai Biju



Accuracy, f1 score, precision, fraud detection, credit card, Random Forest, ADA boost, CAT boost, XGA boost, Light GBM model


: Because of technological advancements,the
growth of Ecommerce industry has been increasedand
it leads to the usage of credit card transactions for
online purchases. Nowadays peoples are most
commonly using online transactions, and it is very
comfortable and helpful. The most prevalent payment
option nowadays is credit card transactions. As a result of
these online transactions, the number of fraud casesis
increasing day by day. Thus, it become one of the great
challenges for the banks to detect this fraud in
transactions. The objective of this work is to find the
fraud in credit card accurately. The machine learning
helps us to detect these type of fraud activities that
occur in credit card transactions in accurate manner. In
this paper we have been included the problems that
causes and activities of fraud in credit card. Several
machine learning algorithms are built by applying
boosting techniques to it, which includes logistic
regression and random forest utilizing ensemble
classifiers on an unbalanced dataset. The algorithms
that have been used in this system are random forest
classifier and ADA boost algorithm. The both of these
algorithms are based on accuracy, precision, and area
under curve score. We compare the results that
obtained from both these algorithms and choose the
one with the greatest accuracy, precision and area
under curve score. Based on this we select the best
algorithm for detecting fraud in credit card
transactions. The conclusion of this research shows
how to use supervised techniques to train and analyse
the best classifier, resulting in a more accurate



How to Cite

Ms. Merin Manoj, & Akshai Biju. (2023). Credit Card Fraud Detection Using Machine Learning. National Conference on Emerging Computer Applications, 4(1).

Most read articles by the same author(s)