Credit card fraud detection using a hierarchical behavior-knowledge space model

Abstract

With the advancement in machine learning, researchers continue to devise and implement effective intelligent methods for fraud detection in the financial sector. Indeed, credit card fraud leads to billions of dollars in losses for merchants every year. In this paper, a multi-classifier framework is designed to address the challenges of credit card fraud detections. An ensemble model with multiple machine learning classification algorithms is designed, in which the Behavior-Knowledge Space (BKS) is leveraged to combine the predictions from multiple classifiers. To ascertain the effectiveness of the developed ensemble model, publicly available data sets as well as real financial records are employed for performance evaluations. Through statistical tests, the results positively indicate the effectiveness of the developed model as compared with the commonly used majority voting method for combination of predictions from multiple classifiers in tackling noisy data classification as well as credit card fraud detection problems.

Credit Card Fraud Detection using Logistic Regression Compared with t-SNE to Improve Accuracy

“…In addition, in Novel Logistic Regression, the average error seems to be higher than t-SNE. It would be preferable if the average error could be considerably reduced (Nandi et al 2022). However, the work can be improved by applying optimization algorithm techniques, to achieve better cross validations and lower mean error (Dorronsoro et al 1997).…”

“…For detection pu. (Nandi et al 2022) Using lost or stolen cards, creating fake or counterfeit cards, copying the original website, deleting or modifying the magnetic strip on the card that holds the user's information, and phishing via skimming or stealing data from a merchant's end are all examples of fraud.. (Nandi et al 2022) Using a credit card is one of the methods for purchasing goods or services. The process of discriminating between fraudulent and non-fraudulent transactions so that customers can enjoy their shopping or other transactions without delay is known as fraud detection.…”

“…A study by (Reddy, Vijaya Kumar Reddy, and Ravi Babu 2018) compares the accuracy of various mining detecting algorithms in detecting credit card fraud. It is important to analyze and compare the various classification algorithms that provide better accuracy (Bonney 1992;Nandi et al 2022). Hence the aim of this research study is to use the Novel Logistic Regression algorithm inorder detect Credit card fraud by achieving better accuracy.…”

Understanding the Offline Criminal Behavior of Individuals Who Live Stream Child Sexual Abuse

“…The volume of cross-border transactions suggests that manual detection of CSA live-streaming transfers is impractical; for this reason machine learning techniques are increasingly applied to financial transactions data for the identification of potential criminal offenses. For example, fraud (OECD, 2021; Nandi et al, 2022), tax avoidance (Korsell, 2015), money laundering, and terrorism financing (Canhoto, 2020) are common applications. However, to detect these offense types, the characteristics of transactions, and offenders, must first be known.…”

A soft voting ensemble learning approach for credit card fraud detection

“…Each model in both phases was evaluated based on the AUROC curve, accuracy, recovery, precision, and F1 score. Padhi et al [ 54 ] implemented six boosting techniques, that is, XGBoost, AdaBoost, Gradient Boosting, LightGBM, CatBoost, and Histogram-based Gradient Boosting, which were hybridized using a stacking framework to predict stock market direction across various datasets from different countries. Employing overfitting protection and evaluating with multiple metrics, the study suggests Meta-LightGBM as a promising predictive model with minimal training and testing accuracy differences, potentially offering investors a tool for risk control and short-term, sustainable profits.…”