Literature Review of Credit Card Fraud Detection with Machine Learning
Rodríguez Vaquero, Patricia (2023)
2023
Master's Programme in Computing Sciences
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-11-10
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202311099537
https://urn.fi/URN:NBN:fi:tuni-202311099537
Tiivistelmä
This thesis presents a comprehensive examination of the field of credit card fraud detection, aiming to offer a thorough understanding of its evolution and nuances. Through a synthesis of various studies, methodologies, and technologies, this research strives to provide a holistic perspective on the subject, shedding light on both its strengths and limitations.
In the realm of credit card fraud detection, a range of methods and combinations have been explored to enhance effectiveness. This research reviews several noteworthy approaches, including Genetic Algorithms (GA) coupled with Random Forest (GA-RF), Decision Trees (GA-DT), and Artificial Neural Networks (GA-ANN). Additionally, the study delves into outlier score definitions, considering different levels of granularity, and their integration into a supervised framework. Moreover, it discusses the utilization of Artificial Neural Networks (ANNs) in federated learning and the incorporation of Generative Adversarial Networks (GANs) with Modified Focal Loss and Random Forest as the base machine learning algorithm. These methods, either independently or in combination, represent some of the most recent developments in credit card fraud detection, showcasing their potential to address the evolving landscape of digital financial threats.
The scope of this literature review encompasses a wide range of sources, including research articles, academic papers, and industry reports, spanning multiple disciplines such as computer science, data science, artificial intelligence, and cybersecurity. The review is organized to guide readers through the progression of credit card fraud detection, commencing with foundational concepts and advancing toward the most recent developments.
In today's digital financial landscape, the need for robust defense mechanisms against credit card fraud is undeniable. By critically assessing the existing literature, recognizing emerging trends, and evaluating the effectiveness of various detection methods, this thesis aims to contribute to the knowledge pool within the credit card fraud detection domain. The insights gleaned from this comprehensive review will not only benefit researchers and practitioners but also serve as a roadmap for the enhancement of more adaptive and resilient fraud detection systems.
As the ongoing battle between fraudsters and defenders in the financial realm continues to evolve, a deep understanding of the current landscape becomes an asset. This literature review aspires to equip readers with the insights needed to address the dynamic challenges associated with credit card fraud detection, fostering innovation and resilience in the pursuit of secure and trustworthy financial transactions.
In the realm of credit card fraud detection, a range of methods and combinations have been explored to enhance effectiveness. This research reviews several noteworthy approaches, including Genetic Algorithms (GA) coupled with Random Forest (GA-RF), Decision Trees (GA-DT), and Artificial Neural Networks (GA-ANN). Additionally, the study delves into outlier score definitions, considering different levels of granularity, and their integration into a supervised framework. Moreover, it discusses the utilization of Artificial Neural Networks (ANNs) in federated learning and the incorporation of Generative Adversarial Networks (GANs) with Modified Focal Loss and Random Forest as the base machine learning algorithm. These methods, either independently or in combination, represent some of the most recent developments in credit card fraud detection, showcasing their potential to address the evolving landscape of digital financial threats.
The scope of this literature review encompasses a wide range of sources, including research articles, academic papers, and industry reports, spanning multiple disciplines such as computer science, data science, artificial intelligence, and cybersecurity. The review is organized to guide readers through the progression of credit card fraud detection, commencing with foundational concepts and advancing toward the most recent developments.
In today's digital financial landscape, the need for robust defense mechanisms against credit card fraud is undeniable. By critically assessing the existing literature, recognizing emerging trends, and evaluating the effectiveness of various detection methods, this thesis aims to contribute to the knowledge pool within the credit card fraud detection domain. The insights gleaned from this comprehensive review will not only benefit researchers and practitioners but also serve as a roadmap for the enhancement of more adaptive and resilient fraud detection systems.
As the ongoing battle between fraudsters and defenders in the financial realm continues to evolve, a deep understanding of the current landscape becomes an asset. This literature review aspires to equip readers with the insights needed to address the dynamic challenges associated with credit card fraud detection, fostering innovation and resilience in the pursuit of secure and trustworthy financial transactions.