Note: This paper was presented at the GCU International Knowledge Transfer Conclave (2018), where it won the Best Paper Award and was included in the conference proceedings (ISBN 978-93-86516-46-6) and has been cited once. It was not accepted in peer-reviewed journals and is not intended as groundbreaking ML research. Instead, it represents an early exploration into the intersection of Blockchain and Machine Learning for fraud detection, during the peak of blockchain adoption in financial services. I keep it public as part of my research journey and recognition milestone.

Blockchain and Machine Learning for Fraud Detection: Employing Artificial Intelligence in the Banking Sector

Harnessing Blockchain and Machine Learning for fraud detection.

Published in Conference Proceedings:

GCU INTERNATIONAL KNOWLEDGE TRANSFER CONCLAVE - 2018 (ISBN 978-93-86516-46-6)

Best Paper Award Winner

Abstract

Fraudulent banking operations can cause huge losses to the bank and further affect the economy negatively. What if Blockchain Technology and Machine Learning could be combined to detect suspicious banking activity and stop transactions at the source? That is what this paper aims to do. In this paper, a system is created which consists of the following components:

1. Blockchain: To securely store transaction history.
2. XGBoosted KMeans algorithm: For quick and efficient detection of outliers, which indicate suspicious activity.

Keywords: Machine Learning, Data Science, Artificial Intelligence, Kmeans Clustering, Blockchain.

Dataset

Here's a data map describing each file's correlation.