PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
April 5, 2026Open Access

AI-Based Credit Card Fraud Detection Using Machine Learning

View Full Paper
Ask AI
Bookmark
Share

Authors

ADADARSH DUBEY

Discussion

Loading...

Member takes

Overview

Demonstrates effective fraud detection in financial transactions using AI, suggesting advanced solutions are needed.

Key Points

  • The aim is to develop an AI-based system for detecting credit card fraud that overcomes limitations of traditional methods.
  • Utilized the Kaggle Credit Card Fraud Detection Dataset with over 284,000 transactions.
  • Applied three supervised classification algorithms: Logistic Regression, Random Forest, and Gradient Boosting.
  • Addressed class imbalance using the Synthetic Minority Over-sampling Technique (SMOTE).
  • Implemented an end-to-end pipeline for data ingestion, feature engineering, model training, and deployment.
  • Random Forest classifier achieved an accuracy of 99.96%.
  • Precision was recorded at 98.7%, and recall at 96.2%.
  • F1-score was determined to be 97.4%, with a ROC-AUC score of 0.9985.
  • Demonstrated superior predictive performance of ensemble machine learning methods for fraud detection.

Cite This Study

ADARSH DUBEY (2026) studied this question.

synapsesocial.com/papers/69d1fdf7a79560c99a0a45e0https://doi.org/10.5281/zenodo.19397610
View Full Paper
Ask AI
Bookmark
Share