Abstract:
With the rise of e-commerce and online purchases in the modern age, credit card fraud has
become a severe and growing issue. Such unpleasant practices can impact millions of people
around the world through this identity theft and the loss of money. Crime is a growing threat
with far-reaching consequences to the finance sector. The extraction of information seemed to
be a core job for payment fraud recognition, fraud detection efficiency in card buy-outs has a
significant impact on the measurement strategy for the data set, the choices of the variable and
the techniques of detection used. We'll aim at how Artificial Neural Networks (ANN), Decision
Trees, K-Nearest Neighbors (KNN), Logistic Regression, Support Vector Machines (SVM),
Random Forests, Neural Network Supervised (MLPClassifier), Ridge Classification,
AdaBoost Classification, and Naive Bayes are implemented in this research. Classification
algorithms for highly skewed credit card fraud results. For model understanding, accuracy, f1,
recall, precision, Matthew's correlation coefficient (MCC), confusion matrix, and lime which
will be used to evaluate the execution of these techniques.
