{"title":"K-Fuse: Credit card fraud detection based on a classification method with a priori class partitioning and a novel feature selection strategy","authors":"Mohammed Sabri, Rosanna Verde, Antonio Balzanella","doi":"10.1002/asmb.2868","DOIUrl":null,"url":null,"abstract":"<p>Online transactions have become the dominant and most popular form of online payment in today's digital economy. Due to the growing popularity of e-commerce and the convenience it offers, both consumers and businesses are rapidly adopting online transactions. Notably, credit cards have become one of the most popular and standard online payment methods. However, it should be noted that credit card transactions are not without challenges. In particular, detecting and preventing fraudulent transactions is a major concern of the online payment system. It is difficult to find an effective detection model that can detect the new patterns created by fraudsters, due to the constant evolution of their methods to exploit the vulnerability of current security protocols. These fraud patterns are evolving and may not correspond to existing documented models, leading to a reduction in their identification. In addition, the customer's behavior can affect the model detection as it is susceptible to change based on factors such as economic conditions, trends, and individual circumstances. When consumers deviate from their typical behavior, the model may generate false alerts, thereby reducing its ability to differentiate between legitimate and fraudulent transactions. This article presents a new supervised detection model, called K-Fuse, which introduces an unsupervised phase in order to detect fraud patterns that may correspond to innovative models introduced by fraudsters. K-Fuse is a supervised classification method that fuses three steps consisting of <i>(i)</i>\nunsupervised clustering to identify hidden patterns of transactions in a dataset, <i>(ii)</i> a novel feature selection criterion based on the unsupervised results, and <i>(iii)</i> supervised classification to exploit the results of clustering and feature selection to predict new transactions as fraudulent or legitimate.</p>","PeriodicalId":55495,"journal":{"name":"Applied Stochastic Models in Business and Industry","volume":null,"pages":null},"PeriodicalIF":1.3000,"publicationDate":"2024-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Stochastic Models in Business and Industry","FirstCategoryId":"100","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/asmb.2868","RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"MATHEMATICS, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0
Abstract
Online transactions have become the dominant and most popular form of online payment in today's digital economy. Due to the growing popularity of e-commerce and the convenience it offers, both consumers and businesses are rapidly adopting online transactions. Notably, credit cards have become one of the most popular and standard online payment methods. However, it should be noted that credit card transactions are not without challenges. In particular, detecting and preventing fraudulent transactions is a major concern of the online payment system. It is difficult to find an effective detection model that can detect the new patterns created by fraudsters, due to the constant evolution of their methods to exploit the vulnerability of current security protocols. These fraud patterns are evolving and may not correspond to existing documented models, leading to a reduction in their identification. In addition, the customer's behavior can affect the model detection as it is susceptible to change based on factors such as economic conditions, trends, and individual circumstances. When consumers deviate from their typical behavior, the model may generate false alerts, thereby reducing its ability to differentiate between legitimate and fraudulent transactions. This article presents a new supervised detection model, called K-Fuse, which introduces an unsupervised phase in order to detect fraud patterns that may correspond to innovative models introduced by fraudsters. K-Fuse is a supervised classification method that fuses three steps consisting of (i)
unsupervised clustering to identify hidden patterns of transactions in a dataset, (ii) a novel feature selection criterion based on the unsupervised results, and (iii) supervised classification to exploit the results of clustering and feature selection to predict new transactions as fraudulent or legitimate.
期刊介绍:
ASMBI - Applied Stochastic Models in Business and Industry (formerly Applied Stochastic Models and Data Analysis) was first published in 1985, publishing contributions in the interface between stochastic modelling, data analysis and their applications in business, finance, insurance, management and production. In 2007 ASMBI became the official journal of the International Society for Business and Industrial Statistics (www.isbis.org). The main objective is to publish papers, both technical and practical, presenting new results which solve real-life problems or have great potential in doing so. Mathematical rigour, innovative stochastic modelling and sound applications are the key ingredients of papers to be published, after a very selective review process.
The journal is very open to new ideas, like Data Science and Big Data stemming from problems in business and industry or uncertainty quantification in engineering, as well as more traditional ones, like reliability, quality control, design of experiments, managerial processes, supply chains and inventories, insurance, econometrics, financial modelling (provided the papers are related to real problems). The journal is interested also in papers addressing the effects of business and industrial decisions on the environment, healthcare, social life. State-of-the art computational methods are very welcome as well, when combined with sound applications and innovative models.