Hybrid approaches for detecting credit card fraud

Abstract

As a natural consequence of the multibillion dollar annual losses incurred as a result of credit card fraud, banks attach great importance to credit card fraud detection. In this paper, we proposed the use of six known models, namely, decision tree, random forest, Bayesian network, Naïve Bayes, support vector machine, and K* models, to form an ensemble for the detection of credit card fraud. We focused on the voting mechanisms used by the ensemble and proposed optimistic, pessimistic, and weighted voting strategies. The proposed model is called optimistic, pessimistic, and weighted voting in an ensemble of models. A dataset of real credit card transactions from a leading bank in Turkey was used to evaluate the performance of optimistic, pessimistic, and weighted voting in an ensemble of models. The results showed that the optimistic voting strategy enables the detection of 31.59% of fraudulent transactions with a false alarm rate of only 0.10%, the pessimistic voting strategy detects 93.92% of fraudulent transactions with a false alarm rate of 13.72%, and the weighted voting strategy detects 64.02% of fraudulent transactions with a false alarm rate of 0.75%. Banks can choose among these voting mechanisms depending on their preferred strategies for fraud detection and desired false alarm rates.

Date

“…With the advent of data mining and machine learning, numerous advances have been made, such as the decision-tree based approaches [1][2][3]29]. When it comes to customs administrations (i.e., the target domain of this research), most tax authorities until today are using rule-based methods [17]. Rule-based systems are interpretable and straightforward, but are brittle against any new behaviors and changes, subjective to expert knowledge, and are cumbersome to maintain [16,21].…”

A Novel text2IMG Mechanism of Credit Card Fraud Detection: A Deep Learning Approach

“…Several studies have been performed in which data mining (DM) and machine learning (ML) methods were used to expose credit card fraud (CCF). In these studies, two main types of methods have been developed for the detection and identification of fraudulent transactions, i.e., unsupervised and supervised methods [4]. In a supervised method, classification is performed by an algorithm based on the transactional data record.…”

“…In a supervised method, classification is performed by an algorithm based on the transactional data record. Hidden Markov models, artificial neural networks [5,6], support vector machines, k-nearest neighbors, random forests, and Bayesian belief networks are some state-of-the-art algorithms used as supervised approaches [4].…”

Credit Card Fraud Detection Using Hybrid Machine Learning Algorithms

“…The dataset is highly unbalanced, the positive class (frauds) account for 0.17% of all transactions. Most banks and financial firms use rule-based systems, in which an expert will use historical fraud data to define a set of rules, and a system will raise an alarm if a new transaction match one of the rules [1,2]. The main limitations of this manual process are that it is reactive, lacks flexibility and consistency as well as the fact that it is time-consuming [3].…”