MétaCan
Menu
Back to cohort
Record W4407421575 · doi:10.62754/joe.v4i2.6377

Advanced Machine Learning Approaches for Credit Card Fraud Detection in the USA: A Comprehensive Analysis

2025· article· en· W4407421575 on OpenAlexaff
Mir Mohtasam Hossain Sizan, Anchala Chouksey, Nikhil Rao Tannier, Md Abdullah Al Jobaer, Jasmin Akter, Ashutosh Roy, Mehedi Hasan Ridoy, M Saif Sartaz, Dewan Aminul Islam

Bibliographic record

VenueJournal of Ecohumanism · 2025
Typearticle
Languageen
FieldComputer Science
TopicImbalanced Data Classification Techniques
Canadian institutionsArtificial Intelligence in Medicine (Canada)
Fundersnot available
KeywordsCredit card fraudCredit cardComputer scienceComputer securityMachine learningBusinessData scienceArtificial intelligenceWorld Wide Web

Abstract

fetched live from OpenAlex

Credit card fraud is a financial threat in America, both for financial institutions and for consumers, and it is growing in severity. Traditional fraud detection methods become less effective in countering emerging fraud trends, and for that reason, sophisticated algorithms in machine learning have to be embraced. This research project strived to develop and compare complex algorithms for fraud detection in credit cards in America. With a variety of algorithms including both unsupervised and supervised learning, this study strived towards improving fraud transaction detection rates. This study focuses on real-world credit card transaction datasets from America, offering a robust foundation for comprehending the intricacies of fraud detection in an authentic financial context. Employing actual transaction data, the study aims to replicate and model variation and nuance in fraud and consumer behavior, such that any developed machine learning algorithms will have a basis in real-life realities. For model selection, we deployed several machine learning models, notably Logistic Regression, Random Forest, and XG-Boost Classifier. In evaluating model performance, several key metrics, including Precision, Recall, and the F1-score, were taken into consideration. Random Forest Classifier performed best overall, with relatively high accuracy for fraud prediction, and average recall, with a marginally high level of F1-score. Overall, it can be noticed that Random Forest has the most balanced performance out of the three in fraud detection capabilities, which seems to be a necessity. The integration of real-time fraud prevention with machine learning models is revolutionizing financial institution transaction monitoring. ML models can analyze and process information in real-time, and thus, allow for effective and efficient real-time fraud monitoring. The future of fraud detection holds many exciting avenues for research, most prominently in deep model development. Methods in deep learning, such as recurrent neural networks (RNNs) and convolutional neural networks (CNNs), have been successful in discovering complex structures and sequential relations in transactional information. Another promising avenue for future research is combining AI-powered identity verification with blockchain technology for fraud prevention.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.004
metaresearch head score (Gemma)0.008
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.009
Threshold uncertainty score0.023

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0040.008
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0090.008
Science and technology studies0.0010.000
Scholarly communication0.0020.002
Open science0.0010.001
Research integrity0.0010.002
Insufficient payload (model declined to judge)0.0010.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.054
GPT teacher head0.299
Teacher spread0.245 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations9
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueJournal of EcohumanismSame topicImbalanced Data Classification TechniquesFrench-language works237,207