Comparison of Logistic Regression and Artificial Neural Network for Modelling Credit Card Data Set with the Identification of Precise Fraudulent

Abstract

Using the Logistic Regression technique, estimate the accuracy % of credit card fraudulent transactions. The accuracy percentage of credit card fraudulent transactions was predicted using a novel decision boundary logistic regression with a sample size of 100 and an artificial neural network (ANN) with a sample size of 100, a 95 percent confidence interval, and a pretest power of 80% iterated at various times. The sigmoid function is used in logistic regression to map values between 0 and 1, which aids in improving the accuracy percentage prediction. Logistic regression has a significantly higher accuracy rate (98.2 percent) than ANNs (88.8 percent). With (p=0.001) (p0.005), there was a statistically significant difference between Logistic regression and ANN. The Logistic Regression algorithm demonstrates a higher predictive accuracy percentage for fraudulent credit card transactions.