EUVIMEDEUVIMED entdecken
Crossref

A Comprehensive Ensemble Approach Using Blending and Stacking for Credit Card Fraud Detection

Bharti Chugh, Preeti Garg, Karnika Dwivedi

International Journal on Smart & Sustainable Intelligent Computing · 2024

DOI: 10.63503/j.ijssic.2024.11 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal on Smart & Sustainable Intelligent Computing, 2024. Verfasst von Bharti Chugh und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Detection of credit card fraud transactions is a severe problem, which requires analyzing large volumes of transaction data to identify fraud patterns. It re-quires finding, which transactions are fraudulent out of millions of daily transactions. As the amount of data is increasing, it is now difficult for an indi-vidual to detect meaningful patterns from transaction data, often character-ized by many samples, many dimensions, and online updates. As a result, there is a need for the best possible approach using machine learning that auto-mates the process of identifying fraudulent patterns from large volumes of data. Therefore, this study proposed a comprehensive ensemble approach us-ing Blending (Voting Classifier) and stacking for credit card fraud detection. As the dataset is imbalanced, the proposed method balanced the dataset resampling techniques. Working with selected features instead of all the features reduces the risk of over-fitting, improves accuracy, and decreases the training time. Afterward, three base classifiers with chosen features, an ensemble voting classifier and a Stacking Classifier have been developed. The computational results indicate that the suggested stacking ensemble is the best, and its Random Forest (RF) classifier has also the best performance among other base classifiers. The ensemble stacking classifiers lead to 82.5% recall,85% F1-score and 82.5 % AUC which has superiority over other ensemble-based methods.

Zitieren

APA

Chugh, B., Garg, P., & Dwivedi, K. (2024). A Comprehensive Ensemble Approach Using Blending and Stacking for Credit Card Fraud Detection. International Journal on Smart & Sustainable Intelligent Computing. https://doi.org/10.63503/j.ijssic.2024.11

Vancouver

Chugh B, Garg P, Dwivedi K. A Comprehensive Ensemble Approach Using Blending and Stacking for Credit Card Fraud Detection. International Journal on Smart & Sustainable Intelligent Computing. 2024. doi:10.63503/j.ijssic.2024.11

BibTeX
@article{Chugh2024,
  author = {Chugh, Bharti and Garg, Preeti and Dwivedi, Karnika},
  title = {A Comprehensive Ensemble Approach Using Blending and Stacking for Credit Card Fraud Detection},
  journal = {International Journal on Smart & Sustainable Intelligent Computing},
  year = {2024},
  doi = {10.63503/j.ijssic.2024.11}
}
Kostenlos starten →