Enhancing business security through fraud detection in financial transactions

Tanvir Rahman Akash *, Md Shakil Islam and Md Sultanul Arefin Sourav

Student, Business Analytics, Trine University, Phoenix, Arizona, United States.
 
Research Article
Global Journal of Engineering and Technology Advances, 2024, 21(02), 079–087.
Article DOI: 10.30574/gjeta.2024.21.2.0205
Publication history: 
Received on 22 September 2024; revised on 05 November 2024; accepted on 07 November 2024
 
Abstract: 
Today the threat of fraud is a real issue in security threats for business, due to complexity and volume of financial transactions, especially with the rapid changes of technology. This research paper focuses on the major role that needs to be implemented for the prevention of fraudulent activities regarding monetary issues. The paper also explores different approaches that are currently used in the identification of frauds such as the artificial neural networks, statistical analysis, rules based systems and others. From the existing literature review and case studies, the paper gives an overview of the strengths and weaknesses of these approaches as well as the implementation of the best practices. The methodology entails a literature review of the current fraud detection systems, then using the identified techniques on a sample dataset to test them. In this paper research data shows that combining several detection measures improves efficiency and minimizes the number of false alarms, which contributes to the improvement of business protection. The discussion also underlined the necessity for the constant appropriateness of technologies dedicated to fraud detection to correspond to new fraud schemes. Finally, recommendations regarding occurrence of effective anti-fraud measures for firms are given in order to promote effective financial transactions and safeguard organizational resources.
 
Keywords: 
Fraud Detection; Business Security; Financial Transactions; Machine Learning; Artificial Intelligence; XGBoost; Data Preprocessing and Model Performance Metrics JEL Classification- G20; G21; G28; C80; M15; K22; K4
 
Full text article in PDF: