Sign in Subscribe
Robotics

Enhancing Credit Card Fraud Detection with Machine Learning

Strategic Insights

— 1 min read

Researchers from Shaqra University have made a significant breakthrough in combating credit card fraud with the help of machine learning algorithms. According to a recent study, the team has successfully developed a robust model that can detect fraudulent transactions with an accuracy of 99.95%. This innovative approach combines traditional machine learning techniques with deep learning models, addressing class imbalance and enhancing predictive accuracy.

Key Takeaways:

  • The researchers developed a model that leverages logistic regression, decision tree, and random forest techniques to detect credit card fraud.
  • The study used the Synthetic Minority Over-Sampling Technique (SMOTE) to mitigate class imbalance and hyperparameter tuning to optimize model configurations.
  • The deep learning model, incorporating focal loss, achieved the highest precision and demonstrated its potential in minimizing false positives.
  • The study validated its approach on both the Kaggle credit card dataset and the PaySim synthetic mobile money dataset, demonstrating robustness and cross-domain generalizability.
  • The researchers found that the random forest model achieved the best overall performance, with an accuracy of 99.95% and F1 score of 0.8256.
  • The study highlights the effectiveness of combining data preprocessing, resampling techniques, and model optimization for robust fraud detection.
  • Senior researcher Samia Dardouri, from Shaqra University's Department of Computer Science, noted the importance of addressing class imbalance and enhancing predictive accuracy in detecting credit card fraud.

Statistics:

  • Accuracy: 99.95%
  • F1 score: 0.8256
  • ROC-AUC: 0.9759
  • Precision: highest among all models, demonstrating its potential in minimizing false positives
  • Number of datasets used: 2 (Kaggle credit card dataset and PaySim synthetic mobile money dataset)

Sources:

  • VerticalNews (2025 Nov 3) - [No additional source information provided in the Text]
  • Enhancing credit card fraud detection using traditional and deep learning models with class imbalance mitigation (Frontiers in Artificial Intelligence, 2025;8:1643292)

Read more