COMPARATIVE ANALYSIS OF VARIOUS CREDIT CARD FRAUD DETECTION TECHNIQUES
Keywords:
Credit Card Fraud Detection, Machine Learning,Deep Learning, Imbalanced Datasets, Model Performance Evaluation, Naïve Bayes,Random Forest, Logistic Regression, KNN, CART, SVM, LDA, CNN, DNN, LSTM Ecommerce,Prediction, Classification, BankAbstract
In the era of digital advancement, online transactions and digital payment system, credit cards are appearing as the biggest fraud. For improved banking security and reduce the financial threats, this fraud searching is occurring as the crucial element in this regard. This study is committed to have a look and draw comparison between various programs like Naïve Bayes, Random Forest, Logistic Regression,K-Nearest Neighbor,Classification and Regression Tree,Linear Discriminant Analysis,Deep Neural Network,Long Short Term Memory, Convolutional Neural Network and Support Vector Machine. Some tools are used to check the authenticity and precision of the models by testing and training. This may result the Deep Learning models and DNN proved to be the best and authentic models with the value 0. 9991.But in 2nd Dataset, the authenticity was achieved where DNN worked with the value 0.8166 in DL models which proved the best results. These results demonstrate that Deep Learning models perform well in the aspects like fraud searching and in complex transactions patterns. The findings of this research can assist researchers and financial institutions to choose appropriate fraud detection methods to create a secure online payment system process.
