Fraud detection on prapaid Istanbulkart

The pandemic process has caused radical changes in individuals' daily life practices and created deep effects on their routines. Especially with the increased importance of hygiene and social distancing rules, traditional payment methods have been replaced by faster and contactless solutions. In this context, contactless payments made via credit cards, debit cards, and prepaid cards have become widespread, and many people have found this method safer and more practical due to the lack of physical contact. However, this transformation has also brought along serious security problems. With the increasing prevalence of contactless payment systems, there has been a significant rise in fraud and unauthorized usage cases. This situation causes card users to feel insecure about financial safety, and as a result, irregularities in card usage behaviors emerge. These irregular and unusual usage patterns make it even more difficult to detect fraudulent activities. Moreover, the constantly changing nature of fraudulent methods and the unpredictability of user behavior reduce the effectiveness of traditional detection techniques. For this reason, machine learning and artificial intelligence-based approaches have come to the fore in recent years as alternative and powerful tools for fraud detection. In this study, the previously unexamined prepaid İstanbulkart data set, which is used in public transportation, has been studied. A new hybrid deep learning model has been developed for detecting fraudulent usage within this data set. The performance level of the model was tested by comparing it with the European credit card data set and another hybrid model. The results obtained show that the proposed model works with high accuracy and can be effectively applied to complex real-world data sets. These findings reveal that the model is suitable for practical use and can provide a reliable solution for fraud detection in real-life scenarios.