Improvements in Fraud Prevention using Machine Learning

IRJET, 2020

The speedy growth in E-Commerce industry has led to an aggressive increase in the use of credit cards for online purchases. In recent years, credit card fraud has become a major complication for banks as it has become very difficult for detecting fraud in the credit card system. To overcome this obstacle Machine learning plays an important role in detecting the credit card fraud in the transactions. As in Machine learning the machine is trained and it predicts the output so, to predict the various bank transactions various machine learning algorithms are used. This paper examines the performance of K-nearest neighbors, Decision Tree, Logistic regression and Random forest for credit card fraud detection. The dataset considers fraud transactions as "Class 1" and valid ones as the "Class 0". The data set is highly imbalanced, it has about 0.173% of fraud transactions and the rest are valid transactions. Four different machine learning classification algorithms used here and the task is implemented in Python language. The performance of the algorithm is evaluated by accuracy score, f1-score, precision and recall score.

IAEME PUBLICATION, 2024

Technology is increasing at a very rapid pace and with this growing technology ecommerce and online transactions have also grown up and it mostly contains transactions through credit cards. When people use credit cards more often, the chances of credit card fraud are rising drastically. The most susceptible system to fraud is the credit card system. Customers and financial institutions lose billions of dollars per year due to credit card fraud, and criminals are constantly searching for new ways to commit crimes. As a result, for banks and financial institutions to reduce their losses, fraud detection systems have become critical. This research investigates the output of different types of classification models, including Decision Tree, Random Forest, Support Vector Machines (SVM), Naive Bayes, and K-Nearest Neighbors (KNN). An open data set on credit card transactions from Kaggle was used in this research. Three classifiers namely random forest, KNN, and SVM were used for modelling. The performance of the three machine learning models was compared using Precision, Recall, Accuracy, and F1 measures. It was found that Random Forest and KNN performed equally better than SVM. The knowledge acquired from this research may direct future investigations targeted at enhancing the ability of financial institutions and banks to withstand such fraudulent activities

IJCSIS Vol 17 No 9, September Issue, 2019

Financialfraudisanevergrowingwith far consequences in the financial industry.Thegrowth in E-commerce industry has lead to exponential increaseintheUKofcreditcardsfor online purchases and consequently they have been surge in the fraud relatedtoit.Inrecentyears,forbanks hasbecomeverydifficultfordetecting the fraud in credit card system. Machinelearningplaysavitalrolefor detectingthecreditcardfraudsinthe transactions. For predictive these transactions, banks make use of variousmethodologies,pastdatahas beencollectedandnew featuresare beenusedforenhancingthepredictive power. The performance of fraud detectingincreditcardtransactionsis greatly affected by the sampling approach on dataset,selection of variablesanddetectiontechniquescan beused.Thispaperinvestigatesthe performance of fraud Logistic Regression,Random Forest,Decision Tree.Thedatasetofcreditcardscan be collected from Kaggle and it contains 2,84,808 credit card transactions of an European bank dataset. It considers that fraud transactions as the positive classes andgenuineonesasnegativeclass. Thedatasetishighlyimbalanced.The performance of fraud detection in credit card transactions is greatly affectedbythesamplingapproachon dataset,selection of variables and detection techniques used.A hybrid techniqueofundersamplingandover samplingiscarriedoutontheskewed data.Thethreetechniquesareapplied ontherawandthepreprocesseddata. TheworkisimplementedinPython. Theperformanceofthetechniquesis evaluated based on accuracy, sensitivity, specificity, precision, Matthews’sco-relationcoefficientand balanced classifications. Logistic Regression is best among allthe algorithms. Keywords— Creditcard fraud,Credit Card Fraud detection methods ,ElectronicCommerce,HiddenMarkov Model(HMM),Logistic Regression., Random Forest

Journal of Intelligent Learning Systems and Applications, 2019

Credit card fraud is a wide-ranging issue for financial institutions, involving theft and fraud committed using a payment card. In this paper, we explore the application of linear and nonlinear statistical modeling and machine learning models on real credit card transaction data. The models built are supervised fraud models that attempt to identify which transactions are most likely fraudulent. We discuss the processes of data exploration, data cleaning, variable creation, feature selection, model algorithms, and results. Five different supervised models are explored and compared including logistic regression, neural networks, random forest, boosted tree and support vector machines. The boosted tree model shows the best fraud detection result (FDR = 49.83%) for this particular data set. The resulting model can be utilized in a credit card fraud detection system. A similar model development process can be performed in related business domains such as insurance and telecommunications, to avoid or detect fraudulent activity.

Journal of Computer Science and Technology Studies

The massive development of technology has affected commerce and given rise to e-commerce and online shopping. Nowadays, consumers prioritize e-shopping over the brick and motor stores due to numerous benefits, including time and transport convenience. However, this progressive upsurge in online payment increases the number of credit card frauds. Therefore, defending against fraudsters’ activity is obligatory and can be achieved by securing credit card transactions. The objective of this paper is to build a model for credit card fraud detection using Machine learning techniques. An innovative approach to credit card fraud detection grounded on machine learning is proposed in this study. Machine learning (ML) is an artificial intelligence subfield comprising learning techniques from experience and completing tasks without being explicitly programmed. Three ML techniques have been used: Support vector machine, logistic regression, Random Forest, and Artificial Neural network. First, th...

International Journal for Research in Applied Science & Engineering Technology (IJRASET), 2023

Credit cards are becoming the most widely utilized form of payment. The numbers of fraud users are growing as quickly as the technology. This paper discusses the performance of three popular Machine Leaning techniques for predicting credit card fraud detection. In this paper we used Machine Learning algorithms such as Decision Tree, Random Forest and Simple Regression tree for handling the imbalanced credit dataset. KNIME is used as data analytical tool for the purpose of simulation. The Comparative Study related to Accuracy is being tested and the Regression tree will give the best result.

The era of variety in digital convenience to make customers loyal to brands, brings along with it an empirical universe of financial fraud. In banks and other financial institutions have shown a higher leap of financial fraud over the past year. This paper describes probability of fraudulent transactions in prevalence and context of credit card usage. A conscious effort is put to bring in a conceptual distinction between fraud detection and predicting probable fraudulent opportunities on the digital space of financial transactions. This emerges a new dimension of financial fraud as a complex phenomenon that can take very different forms, depending on the market segments and the actors involved.

International Journal for Research in Applied Science & Engineering Technology (IJRASET), 2022

Credit card fraud is one of the important financial frauds that has caused a huge amount of financial losses more than the past. Several protection mechanisms such as Fraud Prevention Systems (FPSs) are used to combat the current credit card frauds, but these methods are not efficient enough to reduce the impact of these frauds. It is therefore important that credit card companies are able to recognize and prevent fraudulent credit card transactions so that customers are not charged for items that they did not purchase. In this research, we present our approach to predict legitimate or fraudulent transactions on Credit Card Fraud Detection Dataset from Kaggle. This work is based on analysis at two levels using performance evaluation metrics such as Precision, Recall, F1_Score and ROC_AUC. In the first stage of the research, the original imbalanced and skewed dataset was used to train, predict and evaluate the six supervised machine learning classifiers considered in this research including: Extreme Gradient Boosting (XGBoost), Random Forest (RF), K-Nearest Neighbour (KNN), Decision Tree (DT), Logistic Regression (LR), and Naïve Bayes (NB) while the same set of classifiers were also trained, predicted and evaluated with the same dataset but now resampled using SMOTE-Tomek, i.e., a combination of both under-sampling and over-sampling technique to eliminate the imbalanced nature of the dataset during the second stage. The results of the two stages are compared to select the best overall performance. However, the result of the second stage of the experiment where models are trained, tested and evaluated with resampled dataset gave the overall best results where XGBoost, RF and DT have 100% in Precision, Recall, F1_Score and ROC_AUC respectively. While comparing the overall results of our research with all the papers reviewed in Section 2 of this work, it is worth noting that our research achieved the best performance so far where 100% were recorded from three different classifiers in all the four metrics used to evaluate our work.

Utilization of credit cards encourages individuals to buy products online via the Internet. Individuals tend to do much of purchasing online or offline by utilizing the credit card facility provided by the bankers to their customers. Credit cards have turned out to be the most prominent facility available to the people around the globe to encourage paperless trades at an enormous speed. Whenever any such trade happens in exchanges or net marketing by using a paperless framework, it is subjected under high risk of fraudulent transactions due to many pitfalls in the security system of the usage of credit cards on the networks. This paper presents a brief survey of important and basic linear and non-linear machine learning algorithms that are focused to predict the fraudulent transactions by studying the patterns present in the credit card transactional datasets. The authors provide the methodology of Random Forest (RF), Support Vector machine (SVM) and Artificial Neural Network (ANN) classifiers to accurately classify whether a unseen credit card transaction is fraudulent or not.

International Journal of Advanced Networking and Applications

With the increasing popularity of Credit card usage, Credit Card fraud also increases. The number of online payment options has expanded thanks to e-commerce and several other websites, raising the possibility of online fraud. As a result, both people and financial institutions suffer significant losses. This research seeks to detect credit card fraud and make attempts to cut down on it. Financial institutions place a high priority on identifying and stopping fraudulent activity. Fraud prevention and detection are pricey, time-consuming, and labor-intensive processes. Several machinelearning algorithms can be utilized for detection. In order to evaluate past customer transaction information and identify behavioral traits, the study's main goal is to develop and apply a special fraud detection algorithm for simulcasting transaction data. Through the research, try to give a genuine solution to Credit card users and make their transactions secure. This research aims to propose a tr...