Published Paper Details:

Credit card fraud detection is currently occurring on a large scale everywhere. This problem stands as there has been a sharp increase in the online transactions and usage of e-commerce platforms. It is essential that credit card companies are able to identify deceitful credit card transactions so that customers do not suffer from unnecessary expenditure. Such problems can be dealt with Data Science along with Machine Learning. Credit card frauds usually occur when theft of the card is involved for any of the purposes that are not authorized or even when the fraudster is able to extract the credit card information for their own use. The credit card fraud detection system was introduced to detect such fraudulent activities. The aim of the project is to focus mainly on machine learning algorithms. Random forest algorithm, logistic regression and the SVM algorithm are being used. The results procured from the algorithms are based on accuracy, precision, recall, and F1-score. The confusion matrix is the basis for the plotting of the ROC curve. A comparison occurs of the Random Forest, Decision tree, LGBM and the Nearest neighbors algorithms and the algorithm that has the greatest accuracy, precision, recall and F1-score is decided as the best algorithm that is used to detect the fraud