Credit Card Fraud Detection: Advanced Real-time Application of Deep Sequential Models

2019 ◽  
Vol 8 (12) ◽  
pp. 1037-1043
Keyword(s):  
Credit Card ◽  
Financial Institution ◽  
Assembly Model ◽  
Suggested Model ◽  
Machine Learning Methods ◽  
Card Payment ◽  
Sequential Models ◽  
Transactional Data ◽  
Sequential Modeling ◽  
Real Time Application

In contemporary times, the rapid growth of e-commerce technologies has rendered it possible for people to select the most popular items in terms of recommended price, quality and quantity among various services, facilities, shops and stores from all around the world. Intriguingly enough, the ease of doing things has also made it easier for fraudsters to abuse this huge opportunity. As credit card has emerged to be the most popular mode of payment, the fraudulent activities using credit card payment technologies are fast increasing as a result. Therefore, it is inevitable for a financial institution to think of an automatic and fool-proof prevention mechanism to detect and prevent fraudulent activities. Although many works have been done in this area using traditional statistical and machine learning methods, they suffer from deficiencies as most of them have not taken the sequential nature of transactional data into account. In this paper, we propose an assembly model based on sequential modeling of data using deep recurrent neural networks and a novel selection mechanism based on an artificial neural network to detect fraudulent actions. As the experiments demonstrated, our suggested model outperforms the advanced models in all assessment criteria.

A Simple and Secure Credit Card-Based Payment System

2010 ◽  
pp. 834-842
Author(s):  
Chi Po Cheong
Keyword(s):  
Private Information ◽  
Credit Card ◽  
Financial Institution ◽  
Payment System ◽  
The Internet ◽  
Sensitive Information ◽  
Payment Systems ◽  
Credit Card Number ◽  
Card Number ◽  
Card Payment

Credit card is the most popular payment method used in Internet shopping. The idea of credit card payment is to buy first and pay later. The cardholder can pay at the end of the statement cycle or they can pay interest on the outstanding balance. Therefore, there are many credit card-based electronic payment systems (EPSs) that have been developed to facilitate the purchase of goods and services over the Internet such as CyberCash (VeriSign), iKP (Bellare, Garary, Hauser, et al, 1995), SET (Visa and MasterCard, 1997), CCT (Li & Zhange, 2004), and so forth. Usually a credit card-based EPS involves five parties: cardholder, merchant, acquirer bank, issuer bank, and financial institution. Internet is an open system and the communication path between each other is insecure. All communications are potentially open for an eavesdropper to read and modify as they pass between the communicating endpoints. Therefore, the payment information transmitted between the cardholder and the merchant through Internet is dangerous without a secure path. SSL (Zeus Technology, 2000) is a good example to secure the communication channel. Besides the issue of insecure communication, there are a number of factors that each participant must consider. For example, merchant concerns about whether the credit card or the cardholder is genuine. There is no way to know the consumer is a genuine cardholder. As a result, the merchant is incurring the increase in losses due to cardholder disputes and frauds. On the other hand, cardholders are worried about the theft of the privacy or sensitive information such as the credit card number. They don’t want any unauthorized usage of their credit cards and any modification to the transaction amount by a third party. These security issues have deterred many potential consumers from purchasing online. Existing credit card-based EPSs solve the problems in many different ways. Some of them use cryptography mechanisms to protect private information. However, they are very complicated, expensive, and tedious (Xianhau, Yuen, Ling, & Lim, 2001). Some EPSs use the Certificate Authority (CA) model to fulfill the authentication, integrity, and nonrepudiation security schemes. However, each participant requires a digital certificate during the payment cycle. These certificates are issued by independent CAs but the implementation and maintenance cost of this model is very high. In addition, the validation steps of Certificate-based systems are very time-consuming processes. It requires access to an online certificate server during the payment process. Moreover, the certificate revocation list is a major disadvantage of the PKI-based certification model (The Internet Engineering Task Force). The cardholder’s certificate also includes some private information such as the cardholder’s name. The requirement of a cardholder’s certificate means software such as e-Wallet is required to be installed on the cardholder’s computer. It is the barrier for the cardholder to use Certificatebased payment systems. To solve this problem, Visa Company has developed a new payment system called Verified by Visa (VbV) (http:www/visa-asia.com/ ap/sea/merchants/productstech/vbv_implementvbv. shtml). However, sensitive information such as credit card number is still passed to the merchant. Therefore, the cardholder is not protected by the system.

Applying Machine Learning Methods for Credit Card Payment Default Prediction With Cost Savings

2022 ◽  
pp. 285-305
Author(s):  
Siddharth Vinod Jain ◽  
Manoj Jayabalan
Keyword(s):  
Machine Learning ◽  
Credit Risk ◽  
Financial Services ◽  
Credit Card ◽  
Large Scale ◽  
Cost Savings ◽  
Learning Models ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Card Payment

The credit card has been one of the most successful and prevalent financial services being widely used across the globe. However, with the upsurge in credit card holders, banks are facing a challenge from equally increasing payment default cases causing substantial financial damage. This necessitates the importance of sound and effective credit risk management in the banking and financial services industry. Machine learning models are being employed by the industry at a large scale to effectively manage this credit risk. This chapter presents the application of the various machine learning methods like time series models and deep learning models experimented in predicting the credit card payment defaults along with identification of the significant features and the most effective evaluation criteria. This chapter also discusses the challenges and future considerations in predicting credit card payment defaults. The importance of factoring in a cost function to associate with misclassification by the models is also given.

A Simple and Secure Credit Card-Based Payment System

2009 ◽  
pp. 1299-1306
Author(s):  
Chi Po Cheong
Keyword(s):  
Private Information ◽  
Credit Card ◽  
Financial Institution ◽  
Payment System ◽  
The Internet ◽  
Sensitive Information ◽  
Payment Systems ◽  
Credit Card Number ◽  
Card Number ◽  
Card Payment

Credit card is the most popular payment method used in Internet shopping. The idea of credit card payment is to buy first and pay later. The cardholder can pay at the end of the statement cycle or they can pay interest on the outstanding balance. Therefore, there are many credit card-based electronic payment systems (EPSs) that have been developed to facilitate the purchase of goods and services over the Internet such as CyberCash (VeriSign), iKP (Bellare, Garary, Hauser, et al, 1995), SET (Visa and MasterCard, 1997), CCT (Li & Zhange, 2004), and so forth. Usually a credit card-based EPS involves five parties: cardholder, merchant, acquirer bank, issuer bank, and financial institution. Internet is an open system and the communication path between each other is insecure. All communications are potentially open for an eavesdropper to read and modify as they pass between the communicating endpoints. Therefore, the payment information transmitted between the cardholder and the merchant through Internet is dangerous without a secure path. SSL (Zeus Technology, 2000) is a good example to secure the communication channel. Besides the issue of insecure communication, there are a number of factors that each participant must consider. For example, merchant concerns about whether the credit card or the cardholder is genuine. There is no way to know the consumer is a genuine cardholder. As a result, the merchant is incurring the increase in losses due to cardholder disputes and frauds. On the other hand, cardholders are worried about the theft of the privacy or sensitive information such as the credit card number. They don’t want any unauthorized usage of their credit cards and any modification to the transaction amount by a third party. These security issues have deterred many potential consumers from purchasing online. Existing credit card-based EPSs solve the problems in many different ways. Some of them use cryptography mechanisms to protect private information. However, they are very complicated, expensive, and tedious (Xianhau, Yuen, Ling, & Lim, 2001). Some EPSs use the Certificate Authority (CA) model to fulfill the authentication, integrity, and nonrepudiation security schemes. However, each participant requires a digital certificate during the payment cycle. These certificates are issued by independent CAs but the implementation and maintenance cost of this model is very high. In addition, the validation steps of Certificate-based systems are very time-consuming processes. It requires access to an online certificate server during the payment process. Moreover, the certificate revocation list is a major disadvantage of the PKI-based certification model (The Internet Engineering Task Force). The cardholder’s certificate also includes some private information such as the cardholder’s name. The requirement of a cardholder’s certificate means software such as e-Wallet is required to be installed on the cardholder’s computer. It is the barrier for the cardholder to use Certificatebased payment systems. To solve this problem, Visa Company has developed a new payment system called Verified by Visa (VbV) (http:www/visa-asia.com/ ap/sea/merchants/productstech/vbv_implementvbv. shtml). However, sensitive information such as credit card number is still passed to the merchant. Therefore, the cardholder is not protected by the system.

scholarly journals Evidence from mass-transactional data that chaotic spending behaviour precedes consumer financial distress

2020 ◽  
Author(s):  
Naomi Kate Muggleton ◽  
Edika G Quispe-Torreblanca ◽  
David Leake ◽  
John Gathergood ◽  
Neil Stewart
Keyword(s):  
Financial Distress ◽  
Financial Institution ◽  
Ecological Validity ◽  
Psychological Traits ◽  
High Entropy ◽  
Entropy Measures ◽  
Candidate Personality ◽  
Data Source ◽  
Chaotic Nature ◽  
Transactional Data

The prevalence of digital footprints can allow researchers to study the personalities of millions of individuals with improved ecological validity. We present spending entropy as a candidate personality trait derived as a feature of an objective big data source---mass-transactional data from millions of bank accounts. Entropy measures the unpredictability of spending and acts as a measure of the chaotic nature of a person's life. Over and above how much money people spend, and what the money is spent on, spending entropy positively relates to future financial distress. High entropy leads to increased probability of missed payments across financial products. Entropy temporally relates to future distress three months ahead including more severe measures of distress. We replicate our findings in personal current account, loan, and mortgage holders in a second financial institution. Our findings suggest that high-dimensional data can be used to build psychological traits that predict outcomes in novel situations.

On-Line Credit Card Payment Processing and Fraud Prevention for e-Business

2010 ◽  
pp. 455-473 ◽  
Author(s):  
James G. Williams ◽  
Wichian Premchaiswadi
Keyword(s):  
Credit Card ◽  
Preventive Measures ◽  
Personal Information ◽  
Credit Cards ◽  
The Internet ◽  
Payment Methods ◽  
On Line ◽  
Card Payment ◽  
Debit Cards ◽  
Cyber Criminals

As the volume of purchases for products and services on the Internet has increased and the chosen method of payment is a credit or debit card, e-commerce merchants must be capable of accepting such payment methods. Unfortunately, cyber-criminals have found ways to steal personal information found on credit cards and debit cards and fraudulently use this information to purchase products and services which costs merchants lost revenue and fees for chargebacks. This article discusses the process by which credit card payments are processed beginning with the e-commerce merchant’s web site to a credit card processor or service gateway to the credit card company’s network to the issuing bank’s network with an accept or decline response being returned to the merchant’s shopping cart system via the same networks. The article addresses the issue of credit card fraud in terms of how the cyber-criminals function and the potential solutions used to deter these attempts by the cybercriminals. A list of preventive measures that should be used by e-commerce merchants is provided.

Estimation Procedures of Using Five Alternative Machine Learning Methods for Predicting Credit Card Default

2019 ◽  
Vol 22 (03) ◽  
pp. 1950021 ◽  
Author(s):  
Huei-Wen Teng ◽  
Michael Lee
Keyword(s):  
Machine Learning ◽  
Risk Management ◽  
Credit Risk ◽  
Portfolio Management ◽  
Credit Card ◽  
Nearest Neighbors ◽  
Support Vector ◽  
Credit Risk Management ◽  
Learning Methods ◽  
Machine Learning Methods

Machine learning has successful applications in credit risk management, portfolio management, automatic trading, and fraud detection, to name a few, in the domain of finance technology. Reformulating and solving these topics adequately and accurately is problem specific and challenging along with the availability of complex and voluminous data. In credit risk management, one major problem is to predict the default of credit card holders using real dataset. We review five machine learning methods: the [Formula: see text]-nearest neighbors decision trees, boosting, support vector machine, and neural networks, and apply them to the above problem. In addition, we give explicit Python scripts to conduct analysis using a dataset of 29,999 instances with 23 features collected from a major bank in Taiwan, downloadable in the UC Irvine Machine Learning Repository. We show that the decision tree performs best among others in terms of validation curves.

Institutional determinants of financial inclusion: evidence from world economies

2020 ◽  
Vol 19 (2) ◽  
pp. 217-228
Author(s):  
Tarek Eldomiaty ◽  
Rasha Hammam ◽  
Rawan El Bakry
Keyword(s):  
Credit Card ◽  
Financial Institution ◽  
Panel Data Analysis ◽  
Political Stability ◽  
Financial Inclusion ◽  
Content Type ◽  
Global Database ◽  
Institutional Determinants ◽  
World Governance Indicators ◽  
The Impact

Purpose Financial inclusion is an approach for mobilizing saving and facilitating investments that help promote economic development and pave the way for sustainable development. This paper aims to examine the impact of world governance indicators (WGIs) on the improvement of financial inclusion across world economies. Design/methodology/approach This paper uses the global database of financial inclusion indicators (global findex) for the years 2011, 2014 and 2017. The WGIs are used as proxies for the effects of governmental institutional arrangements. Using panel data analysis, a fixed generalized linear model is estimated for four common financial indicators; namely, borrowed from a financial institution, saved at a financial institution, credit card and debit card ownership. Findings The empirical results reveal that control of corruption, government effectiveness, political stability and voice and accountability are the significant WGIs that influence financial inclusion significantly. Originality/value This paper contributes to the literature in two ways. First, this paper offers validating the results previously reported in related studies. Second, this paper offers robust estimates of the effects of the institutional WGIs on the promotion of financial inclusion.

Sign in / Sign up

Export Citation Format

Share Document