Optimization of Credit Card Fraud Based on Roulette and Tournament Selection by Genetic Algorithm and Artificial Bee Colony

Authors : Rıdvan Çubukcuoğulları, Hasan Temurtaş, Çiğdem Bakır
Pages : 54-66
Doi:10.17694/bajece.1447975
View : 93 | Download : 117
Publication Date : 2025-03-30
Article Type : Research Paper
Abstract :Credit card fraud has become a major problem, especially with the spread of e-commerce and the increase in online shopping during the COVID-19 period. Theft of credit card information and transactions made with cards belonging to others lead to legal and economic problems. The aim of this study is to develop Genetic Algorithm (GA) and Artificial Bee Colony (ABC) methods based on roulette and tournament selection instead of random selection in the determination and classification of these illegal credit card fraud transactions. A data set consisting of 28 attributes and 284,807 credit card transactions was used in the study. Fraud in credit card transactions was estimated using genetic algorithm and artificial bee colony and the obtained results were compared. Optimization methods such as GA and ABC developed based on roulette and tournament selection were analyzed separately according to linear, quadratic and exponential functions. Using the linear, quadratic and exponential functions for the test data, the GA based on roulette selection showed success rates of 98.6%, 98.46% and 98.6% in identifying credit card fraudulent transactions, respectively; With the GA based on tournament selection, it shows 98.53%, 98.33% and 98.66% success. In addition, for the test data, using linear, quadratic and exponential functions, ABC method achieved 98.6%, 97.86% and 97.93% success in determining credit card fraudulent transactions, respectively. Success results were calculated with different evaluation criteria such as accuracy, recall and F1-Score and performance evaluation was presented for each proposed method.
Keywords : artificial bee colony, genetic algorithm, linear, roulette, tournament.

ORIGINAL ARTICLE URL