EZE, CHARITY MMESOMA (2026) DEVELOPMENT OF AN AI-POWERED PAYMENT GATEWAY SYSTEM WITH FRAUD DETECTION AND AUTOMATED RECEIPT GENERATION. Other thesis, GODFREY OKOYE UNIVERSITY, ENUGU STATE..
|
Text
PDF_..EZE CHARITY MMESOMA PROJECT ARRANGEDDDD WORK.pdf Download (1MB) | Preview |
Abstract
The development of digital payments in Nigeria has enhanced efficiency in transaction processing; nevertheless, the current payment gateways are characterized by several problems such as absence of fraud detection tools, failure to provide automated receipts, and lack of a proper merchant analysis tool. The purpose of this study was to design and create an AI payment gateway system known as Payvora. This payment gateway system seeks to improve transaction security and the experience of the merchants by ensuring secure payments, instant fraud detection, creation of receipts and intelligent analysis of transactions. This study is very important because it will contribute significantly to making digital finance stronger in Nigeria by creating a smarter and more secure payment process. This was accomplished through the use of machine learning algorithms, in particular the XGBoost algorithm, along with rule-based pre-processing and feature engineering. The fraud detection model was trained on the PaySim dataset, which is an artificial mobile money transactions dataset that aims at simulating financial fraud cases found within African digital financial systems. The model was analyzed using performance measures such as accuracy, precision, recall, F1-measure, confusion matrix. The experimental findings indicated that the model attained 96.4% for accuracy, 94.8% for precision, 95.2% for recall, and an F1-score of 95.0%, which is a clear indication of the model's effectiveness in detecting fraud with minimal false positives. React.js, HTML, CSS, JavaScript, were used for the front-end design of the application, FastAPI with Python was utilized in the back end, and MongoDB was chosen for the database. Paystack and Flutterwave APIs made it possible for real-time processing of payments and webhooks. This research has contributed to knowledge by showing that machine learning can be integrated into payment gateway systems to help improve transaction security, automation of receipt production, merchant analytics, among other aspects.
| Item Type: | Thesis (Other) |
|---|---|
| Subjects: | Q Science > Q Science (General) |
| Divisions: | Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science |
| Depositing User: | Nnenna Ayo |
| Date Deposited: | 22 Jul 2026 09:50 |
| Last Modified: | 22 Jul 2026 09:50 |
| URI: | http://eprints.gouni.edu.ng/id/eprint/5965 |
Actions (login required)
 |
View Item |
