Fraud Net Detection System
Keywords:
AI, ML, SVM, fraud detectionAbstract
Fraud detection is essential across industries such as finance, insurance, e-commerce, and healthcare. Traditional methods, which rely on rule-based systems and manual oversight, often fail to detect new and evolving fraudulent patterns. By automating the procedure and increasing accuracy, the combination of artificial intelligence (AI) and machine learning (ML) has greatly enhanced fraud detection. In order to find intricate patterns and anomalies, this study examines AI-driven methods that analyse massive amounts of transaction data in real time, such as neural networks, decision trees, support vector machines (SVM), and deep learning models. AI models are extremely flexible and constantly increase their accuracy by learning from fresh data. However, challenges such as data privacy, model interpretability, and adversarial attacks remain. This paper discusses various AI-based fraud detection models, their effectiveness, and the challenges faced, concluding that AI provides a scalable and efficient solution to combat fraud.
DOI: https://doi.org/10.24321/2456.1428.202533
References
Carcillo, F., Laurent, A., Le Borgne, Y. A., Caelen, O., Bontempi, G., & Cordier, M. (2016). "Combining unsupervised and supervised learning in credit card fraud detection." Information Sciences, 345, 26-47.
Feng, X., Xu, B., Li, X., & Li, Y. (2017). "Credit card fraud detection based on SVM and grid search." International Journal of Security and Its Applications, 11(3), 1-12.
Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2018). "SMOTE: Synthetic Minority Over-sampling Technique." Journal of Artificial Intelligence Research, 16, 321-357.
