Scribd
Upload
32 views8 pages

Bank Fraud Detection System Using Machine Learning

Uploaded by

aaaarrvind
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
32 views8 pages

Bank Fraud Detection System Using Machine Learning

Uploaded by

aaaarrvind
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 8

Bank Fraud Detection System

using Machine Learning


Introduction

 Theproject focuses on developing a bank


fraud detection system for Bank XYZ using
machine learning techniques to enhance
financial security, reduce fraud-related
losses, and build customer trust.
Motivation

 Significance:Fraud costs financial institutions


millions of dollars annually.
 Impact:Fraud can erode customer trust and
result in substantial financial losses.
 Real-WorldExample:In 2023, global fraud
losses reached $32 billion. An AI-powered
system can help reduce this number.
Literature Review

Study on Banking Frauds Scenario in India:


 The Reserve Bank of India (RBI) reported a 28%
increase in fraud cases by volume and a 159% rise in
the associated financial losses between 2018-2020​
 Advances frauds accounted for 53% of total fraud
cases in 2019-2020, with the amount involved rising
significantly​.
 RBI introduced new frameworks for fraud detection
and reporting to combat these challenges​.
Literature Review

Analysis and Preventive Measures of Banking


Frauds:
 Thestudy highlighted that private sector banks
experienced the highest percentage of frauds .
 Advances and internet frauds were the most
common, accounting for 41.45% and 40.03% of
cases, respectively​.
 Implementing AI and forensic auditing is
suggested to enhance fraud detection​.
Proposed Solution
Approach:
 Use machine learning algorithms (e.g., Random Forest, SVM) for dynamic fraud
detection, utilizing past transaction data to identify suspicious activities in real
time.
Key Features:
 Real-time transaction monitoring, anomaly detection, and automated fraud
alerts.
Technologies Used:
 Machine Learning Tools: Scikit-learn, TensorFlow
 Languages/Tools: Python (pandas, numpy), Matplotlib/Seaborn for
visualizationHardware
 Requirements: A high-performance system with 16GB RAM and GPU support for
fast model training.
Expected Outcomes / Deliverables

• Deliverables:
• A functional bank fraud detection system using machine
learning.
• Models capable of real-time fraud detection.
• A detailed report and documentation of the system.
• Input:
• Transaction data (e.g., amount, location, time).
• Output:
• Fraud detection alerts with real-time notifications to the user
and analysts.
References

 Baxi, M.K., & Swadia, B.U. (2021). Study on Banking Frauds Scenario
in India. Journal of the Maharaja Sayajirao University of Baroda, 55(1),
56-60. Retrieved from ResearchGate.
 Thunga, S.R., Lavanya, P.B., & Chinni, N. (2023). Analysis and
Preventive Measures of Banking Frauds with Reference to Indian
Banking Sector. Proceedings of the 3rd International Conference on
Reinventing Business Practices, Advances in Economics, Business

You might also like