Development of a Credıt Card Fraud Detectıon System Usıng Artıfıcıal Neutral Network and Support Vector Machıne

  • H. T. DIDIGWU Federal University of Petroleum Resources, Effurun, Delta State
  • R. E. AKO Federal University of Petroleum Resources, Effurun, Delta State
  • S. NIEMOGHA Federal University of Petroleum Resources, Effurun, Delta State
  • M. ASUOBITE Federal University of Petroleum Resources, Effurun, Delta State
Keywords: Artificial Intelligence, Algorithm, Fraud, Machine learning, Detection

Abstract

Credit card fraud has without hesitation an expression of criminal intent and
deception. Fraud identification seems to be a complicated problem that requires a
significant amount of skill until the emergence of machine learning algorithms
were deployed for their classification and detection. However, it is an
implementation for both the better of machine learning as well as artificial
intelligence, ensuring that perhaps the funds of both the customer seems to be
secure and therefore not manipulated. This project therefore proposed the
development of an improved credit card fraud detection system using machine
learning algorithms. The proposed model deployed a fusion of support vector
machine and artificial neural network algorithms for the classification and
detection of credit card fraud was developed using feature driven development
methodology which deploys a feature centric approach to program development
combining different requirement components to meet user needs. The system was
trained and tested with credit card fraud datasets from Kaggle machine learning
repository split into 70:30 ratio for training and testing purposes respectively.
After 20 epochs the model performance outperformed the existing system with an
accuracy of 99.83%, precision 100%, recall 100% and F1-score 100% respectively.

Author Biographies

H. T. DIDIGWU, Federal University of Petroleum Resources, Effurun, Delta State

Computer Science

R. E. AKO, Federal University of Petroleum Resources, Effurun, Delta State

Computer Science

S. NIEMOGHA, Federal University of Petroleum Resources, Effurun, Delta State

Computer Science

M. ASUOBITE, Federal University of Petroleum Resources, Effurun, Delta State

Computer Science

Published
2025-10-01