A Predictive Analytical Study on Detecting Fraudulent Transactions in E-Banking

Sachin A. S; S. Nithyasri; Tejas M; Prajwal R

A Predictive Analytical Study on Detecting Fraudulent Transactions in E-Banking

Author(s)	Mr. Sachin A. S, Ms. S. Nithyasri, Mr. Tejas M, Mr. Prajwal R
Country	India
Abstract	With the massive shift toward digital and mobile banking, the methods used by criminals to steal money have become increasingly advanced. Traditional security measures, which usually rely on simple "if-this-then-that" rules, are no longer enough to stop modern hackers. This study looks into how we can use predictive analytics—like machine learning and deep learning—to build a more proactive defence. The real issue here is that fraud isn't usually a single, glaring event; it's tucked away in underlying patterns that get lost among millions of ordinary, everyday purchases. This research has analysed different models, like Hidden Markov Models and Neural Networks, handle the "imbalanced data" problem. This is a huge issue in banking because 99% of transactions are legitimate, making the 1% of fraud very hard for a computer to "learn" how to spot. By looking at a user’s historical spending habits and behaviour, these predictive tools can flag a transaction as suspicious the moment it happens. It also looks at the trade-offs involved—while deep learning is incredibly accurate at catching tricky scams, it requires a lot of computing power and can sometimes be a "black box," making it hard for bank staff to explain why a transaction was blocked. Ultimately, this study shows that the best approach for e-banking security isn’t just one single tool, but a hybrid system. By combining behavioural analysis with real-time AI monitoring, banks can stay one step ahead of evolving fraud tactics like phishing and identity theft. The goal is to create a system that is fast enough to protect the bank but smooth enough that it doesn’t annoy regular customers with constant "false alarms." This paper highlights that while the technology is getting better, we still need to balance technical accuracy with the practical costs of running these systems in a real-world banking environment.
Keywords	Predictive Analytics, E-Banking Fraud, Machine Learning, Anomaly Detection, Class Imbalance
Field	Business Administration
Published In	Volume 8, Issue 5, September-October 2026
Published On	2026-09-29

View / Download PDF File

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

A Predictive Analytical Study on Detecting Fraudulent Transactions in E-Banking

Share this