International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | UPI Fraud Detection System Using Convolutional Neural Network (CNN) |
|---|---|
| ABSTRACT | The rapid growth of the Unified Payments Interface (UPI) has made digital transactions fast and convenient. However, the increasing use of digital payments has also created opportunities for fraudulent activities such as phishing, impersonation, and misuse of user-behavior patterns. Conventional fraud detection approaches that depend on fixed rules and thresholds may not identify changing or complex transaction patterns effectively. This paper presents UPI Guard, a UPI fraud detection system that uses a Convolutional Neural Network (CNN) to classify transaction activity as valid or fraudulent. The system is developed as a Flask-based web application and includes OTP-based login, role-based portals for users, merchants, and administrators, merchant QR-code generation, and a QR-based payment simulation. During payment initiation, transaction information is represented using pre-scaled features and passed to the CNN model for prediction. The application displays the result as “VALID” or “FRAUDULENT”; transactions identified as suspicious are intended to be blocked, and the user is notified. Transaction records are maintained for administrative monitoring. The proposed system aims to support proactive fraud detection within the project’s simulated UPI payment workflow. The project presentation does not provide dataset details or measured evaluation values; therefore, no numerical performance claims are made in this paper. |
| AUTHOR | MISBA TABASSUM, SHAHEEN MAGI, NAZIYA BANU, MEHEK S, PROF. MANJUNATH K. G. UG Students, Dept. of CSE, Jain Institute of Technology, Davangere, Karnataka, India Assistant Professor, Dept. of Computer Networks & Communication, Jain Institute of Technology, Davangere, Karnataka, India |
| VOLUME | 189 |
| DOI | DOI: 10.15680/IJIRCCE.2026.1410050 |
| pdf/50_UPI Fraud Detection System Using Convolutional Neural Network (CNN).pdf | |
| KEYWORDS | |
| References |
[1] Liu et al., “Fraud Identification Using Machine Learning,” 2025. (Full publication details were not supplied in the project presentation; verify the title, publication venue, pages and DOI.) [2] Patel et al., “Fraud Detection using Random Forest,” 2024. (Full publication details were not supplied in the project presentation; verify the title, publication venue, pages and DOI.) [3] Sharma et al., “ML-Based Financial Fraud Detection,” 2023. (Full publication details were not supplied in the project presentation; verify the title, publication venue, pages and DOI.) [4] Python Software Foundation, “Python Documentation.” https://docs.python.org/ [5] Pallets Projects, “Flask Documentation.” https://flask.palletsprojects.com/ [6] TensorFlow Team, “TensorFlow Documentation.” https://www.tensorflow.org/ |