Credit Card Fraud Detection using Machine Learning and Deployment of Model in Public Cloud as a Web Service
S Kiruthika1, Sowmyarani C N2
1S. Kiruthika*, Department of Computer Science and Engineering, R V College of Engineering, Bengaluru, India.
2Dr. Sowmyarani C.N., Department of Computer Science and Engineering, R V College of Engineering, Bengaluru, India.
Manuscript received on May 25, 2020. | Revised Manuscript received on June 29, 2020. | Manuscript published on July 30, 2020. | PP: 548-552 | Volume-9 Issue-2, July 2020. | Retrieval Number: B3800079220/2020©BEIESP | DOI: 10.35940/ijrte.B3800.079220
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In recent times, usage of credit cards has increased exponentially which has given way to an increase in the number of cybercrimes related to transactions using credit cards. In this paper, the aim is to reduce the fraudulent credit card transactions happening around the world. Latest technologies like machine learning algorithms, cloud computing and web service implementation has been used in this paper. The model uses Local outlier factor algorithm and Isolation forest algorithm to develop the credit card fraud detection model using unsupervised learning techniques. The model has been implemented as a Web service to make the solution integratable with other applications and clients across the world. A third party prototype application is developed and integrated to the Fraud Detection Model using Web Services. The complete Fraud Detection System is deployed on the cloud. The Fraud Detection Model shows exceptionally high accuracy when compared to other models already existing.
Keywords: Fraud detection model, machine learning, local outlier factor, isolation forest, web service, prototype application, public cloud, amazon web services, EC2 instance.