A Comparison of Logistic Regression Classifier and Random Forest Classifier for the Accurate Classification of Credit Card Fraudulent Transactions

Abstract

Aim: The Aim of the research work is credit card fraud detection using random forest classifier in comparison with logistic regression.Materials and Methods:novel random forest classifier algorithm with sample size =09 and novel logistic regression algorithm with sample size =09 were evaluated many times to predict the efficiency percentage.The G  power taken as 0.8 and a =0.05.Results and Discussion:Random forest classifer has been efficiency(97.12%) when compared to logistic regression algorithm efficiency(95.34%). The statistical significance difference(two-tailed) is 0.001(p<0.05).Based on the above it is observed that the Random forest classifier  with  97.12% has better accuracy than 95.34% Logistic regression in Credit card fraud detection.There is a statistical 2-tailed significant difference in accuracy for algorithms is 0.02(p<0.05) by independent t-test.Conclusion:Random forest classifier algorithm significantly better than logistic regression algorithm.