Jul 2026· International Journal of Grid Computing & Applications· Vol 17, pp. 49-61· 0 citations· 28 references
TL;DR
It is found that using more advanced fraud detection systems can reduce financial losses and increase customer trust, which supports overall economic stability and highlights the need to tailor solutions to specific regional and operational conditions.
Abstract
One of the major fraud risks that financial institutions and consumers around the world face is credit card Fraud. This now requires major advanced methods to identify and prevent this from reoccurring. This study compares three main approaches to fraud detection: machine learning, traditional statistical methods, and a combination of both. This research will show how each method works, the cost implications of each method, and the threats and problems that are faced when setting these methods up. We will be highlighting real examples from banks and fin tech companies in the United States, Europe, and West Africa, the research will take us through the day to day standing operating processes. It also points out how results and adoption differ across regions. The findings go to show that machine learning and hybrid approaches generally do a better job at spotting fraud and adapting to new patterns than traditional statistical techniques. However, putting these advanced systems in place is not always easy, as there are issues such as poor data quality, strict regulations, and limited resources, especially in developing regions. This often slows down implementation. The study also finds that using more advanced fraud detection systems can reduce financial losses and increase customer trust, which supports overall economic stability. At the same time, challenges remain, including constantly changing fraud tactics, difficulties integrating new systems with older technology, and a shortage of skilled professionals. Overall, this research provides practical guidance for financial institutions look ing to improve their fraud detection systems and highlights the need to tailor solutions to specific regional and operational conditions.
The fact that banks may find fraud, minimize risks before they happen, and make their clients happy by combining ML and CRM standard data models together in an effective manner is demonstrated.
Satyendra Kumar Vanapalli· International Journal of Mac...· 0 citations
Credit card fraud has become a central issue for the financial security of banks and cardholders. However, fraudulent transactions account for only a very small proportion of all transaction cases. If it is missed, it may directly cause huge property damage. As a result, this study focuses on how to identify fraudulent transactions as much as possible, while controlling false positives and human audit costs. Specifically, logistic regression, gradient boosting, XGBoost, and random forest are compared, while class weighting, SMOTE, and Random Undersampling are evaluated for handling class imbalance. It is more critical to translate the model results into a risk warning system and a loss simulation system. The results show that random forest with class weighting has the best overall performance, with 90.6% F1-Score, 94.1% precision, and 87.3% recall. Meanwhile, under the cost assumptions, this model reduces simulation costs by 85.4% relative to the no-model baseline. These findings suggest that in data environments with severe class imbalances, a machine learning model can be designed as an effective triage tool. Its actual value is not only to predict fraud, but also to establish a low-risk release, medium-risk verification, and high-risk manual review of the decision-making process for institutions and truly reduce losses.
Zi-Yue Meng· Advances in Economics, Manag...· 0 citations
Purpose: Public sector financial systems process large volumes of payments, procurement transactions and grants and remain exposed to fraud, corruption and error. Traditional audit-based controls detect only a small share of irregular transactions and usually do so late. This study examines whether machine learning can improve detection of fraudulent transactions in public sector financial systems and compares model families under conditions resembling real government data, including severe class imbalance and scarce fraud labels. Design/methodology/approach: The study builds a synthetic public-sector transaction dataset of 60,000 records that reproduces documented properties of government payment data, including low fraud prevalence, vendor concentration and period-end clustering. Four models are trained and compared: logistic regression, random forest, XGBoost and isolation forest. Models are evaluated using precision, recall, F1 score, area under the receiver operating characteristic curve and area under the precision-recall curve, since accuracy alone misleads under imbalance. Findings: XGBoost achieved the strongest balance of precision and recall (F1 = 0.762, AUC-ROC = 0.992), followed by random forest (F1 = 0.748). Logistic regression achieved high recall but low precision, producing many false alerts. Isolation forest, which is an unsupervised method, performed reasonably well without labelled fraud, which matters since most agencies lack reliable labels. Amount, vendor age, single-bidder procurement and prior irregularity counts were the strongest predictors. Originality/value: This study contributes a reproducible, coded evaluation framework for public sector fraud analytics and gives practitioners evidence-based guidance on model choice under label scarcity and class imbalance, which are conditions that dominate real government settings.
Keywords: Fraud Detection, Machine Learning, Public Sector, Financial Systems, Public Procurement, Anomaly Detection, Government Auditing, XGBoost.
Emmanuel Afriyie Okyem, Robert Kwasi Aboagye Mensah Yeboah, T. Adukpo· Finance & Accounting Res...· 0 citations
This paper describes a system for the detection of fraud, which is both dynamic and adaptable and which is obtained through the synthesis of machine learning techniques and the CRM data streams and shows how this unified method can lead to an increase in detection performance, shortening of the time for the response, and higher customer confidence in comparison to the existing systems.
Satyendra Kumar Vanapalli· International Journal of Mac...· 0 citations
Now the days world become digital, credit card customers have become common; it makes the payment hassle-free. With the ease of use of credit cards; Fraudulent use of credit cards is growing as a significant affair for financial institutions and consumers on an international level. Traditional rule-based detection algorithms are ineffective in determining a transaction's fraudulent nature. First and foremost, it is imperative to comprehend the pattern of fraudulent activities. The current study explores various supervised machine-learning algorithms to analyze patterns and predict the fraudulent nature of transactions in a large dataset used for training the model. The effectiveness of different methods is assessed by comparing their accuracy, precision, F1score, and recall. In the present paper, we discuss the techniques named KNN, SVM(Support Vector Machine), Logistic Regression, Gradient Boosting, Neural Network, XG Boost, Naïve Bayes, Ada Boost, Decision Forest, and Random Forest.
S. Bansal, Reena Hooda, Rohit Yadav· Journal of Commerce, Economi...· 0 citations
Real-world credit card fraud detection is the primary emphasis of the project. Credit card fraud has lately increased dramatically as a result of the amazing surge in the number of transactions. The goal is to get something without paying for it or to get money out of a bank account without authorization. All credit card issuers must now have effective fraud detection systems in order to reduce their losses. Making the business is a major difficulty since no cardholder or card must be present for a transaction to be completed.. Merchants are unable to determine if a consumer presenting their card is in fact the legitimate owner. It is possible to increase the accuracy of fraud detection by using the suggested technique, which makes use of a random forest. The random forest technique is used to analyse a data set and the current dataset of the user. Finally, improve the precision of the output data. The accuracy, sensitivity, specificity, and precision of the procedures are assessed. Processed characteristics are used to identify fraud, and a graphical model depiction is presented. The accuracy, sensitivity, specificity, and precision of the procedures are assessed.
Devraj Mani, M. Tirupathamma· Journal of Science & Technol...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
