INTERNATIONAL RESEARCH JOURNAL OF MODERNIZATION IN ENGINEERING TECHNOLOGY AND SCIENCE

Abstract

ABSTRACT : With the rapid advancement of electronic commerce technology, there has been a significant increase in the usage of credit cards. As credit cards have become the most popular mode of payment, the occurrence of associated fraud cases has also seen a rise. This paper presents a comprehensive review of existing techniques for detecting credit card fraud. Fraud detection aims to swiftly identify fraudulent activities as they occur. Methods for fraud detection are continually evolving to counter criminals who adapt their strategies. Transactions are categorized as normal, abnormal, or suspicious based on initial assessments. If a transaction is flagged as suspicious, its likelihood of being fraudulent or genuine is further evaluated using Bayesian learning techniques.INTRODUCTION : The prevalence of online shopping has been steadily increasing. A study by ACNielsen in 2005 revealed that one-tenth of the global population engages in online shopping. Credit cards have emerged as the preferred method of payment. With the expanding user base of credit cards worldwide, incidents of identity theft and fraudulent activities have surged. Credit card purchases can be classified into two categories: physical card purchases and virtual card purchases. In a physical card purchase, the cardholder personally presents the card for payment, requiring the attacker to physically steal the card and forge the signature to complete a transaction. Conversely, virtual card purchases only require the card information, such as the card number, expiration date, and secure code, typically used for online or telephone transactions. Committing fraud in virtual purchases simply necessitates knowledge of the card details. Online purchases are predominantly made using credit cards, leading to a rise in credit card fraud incidents. As credit card usage becomes more widespread, financial losses due to fraud also increase. Security measures are crucial for safe credit card usage and prevention of fraudulent activities. The primary aim of security is to prevent unauthorized use of credit cards. Instances of credit card fraud include issues such as lost or stolen cards, application fraud, counterfeit fraud, mail-order fraud, and non-received item (NRI) fraud. Implementing robust security measures for credit cards is essential to mitigate these fraudulent activities. Maintaining the confidentiality of credit card information is essential. To safeguard credit card privacy, it is imperative to prevent any leakage of details. Various methods used to obtain credit card details include phishing websites, theft or loss of physical cards, counterfeit cards, card detail theft, and interception of cards. To enhance security, it is crucial to mitigate these risks. Credit card security measures are vital for distinguishing between valid and invalid transactions. It's important to note that many fraudulent transactions stem from the use of stolen card numbers rather than the physical theft of the card itself. Therefore, it is essential to handle and store credit cards securely. Internet-based frauds, such as online credit card scams, are gaining popularity due to their inherent characteristics. In these online frauds, transactions are conducted remotely, requiring only the card details without the need for manual signatures, PINs, or card imprints 1 during purchase. Often, the legitimate cardholder remains unaware that their card information has been viewed or stolen by someone else. An effective method for detecting this type of fraud involves analyzing spending patterns for each card and identifying any deviations from the usual spending patterns. Analyzing the historical purchase data of cardholders is considered the most effective method for reducing the occurrence of successful credit card frauds. Transactions conducted using payment cards, including credit cards, prepaid cards, debit cards, and smartphones, are classified as fraudulent if unauthorized. Credit card fraud is characterized as unauthorized account activity by individuals not intended to use the account. Operationally, this refers to an event where action can be taken to halt ongoing abuse and implement risk management measures to prevent similar incidents in the future. Credit card fraud occurs when an individual uses someone else's credit card without the knowledge of both the cardholder and the card issuer. Methods for detecting fraud are devised to prevent such unlawful activities by offenders. However, the development of new fraud detection techniques is hindered by the scarcity of innovative ideas in this field, exacerbated by limited access to datasets and undisclosed results. Detecting fraudulent cases relies on available datasets, often referred to as logged data, and user behavior analysis. Currently, various methods, including data mining, statistical analysis, and artificial intelligence, are utilized for fraud detection.

Paper File to download :