How Name Matching Helps Protect Against Credit Card Fraud

Credit Card Fraud is a Global and Steadily Growing Problem

Credit Card Fraud occurs when a credit card is used for a fraudulent charge. Typically, fraudsters employ stolen cards or engage in identity theft. For types of credit card fraud, see here.

Motley Fool reports that credit card fraud has been on a steady rise in recent years. There were more than 180,000 more cases reported in 2024 than in 2019. A disproportionate amount occurs in the United States.

The majority of credit card fraud occurs in card-not-present (CNP) transactions, where the payment card is not physically presented during the transaction, such as with online transactions. With the rise of remote payment scenarios such as e-commerce, online shopping, phone orders, and mail orders, there are more and more opportunities for this type of credit card fraud.

Typically, fraudsters obtain stolen credit cards through various means like data breaches, the dark web, phishing, and skimming.

The financial burden of credit card fraud often falls on the issuing banks but sometimes on the merchants if fraud occurred due to outdated technology or poor security. If a merchant is held liable, they may incur not only the loss of the sale, but also chargeback fees, penalties, and administrative costs.

How Fuzzy Matching Helps Verify that a Transaction is Legitimate

One main way to protect against credit card fraud is to verify that the name on the transaction matches the cardholder’s name.

Merchants and credit card companies can—upon request for verification by the merchant or some other originator—compare the name given by the individual during the transaction against the cardholder’s name on record. Other information is also typically provided for verification, such as the PAN (Primary Account Number), the expiration date, the billing address, and the CVV2 number (a three or four-digit security code found on the back of most credit and debit cards).

A big challenge for verification is that the name provided by the consumer can legitimately be different from the name on record due to many factors (more on this just below).

Consequently, issuing banks, credit card companies, and/or merchants need to make sure that the name is sufficiently similar to the name on record before approving a transaction.

Why Matching Names Is Challenging

As alluded to above, person names can vary in many different and innocent ways, among which are:

• • Nicknames: Ellen Jones vs. Ellie Jones vs. El Jones vs. Ella Jones vs. Elena Jones vs. Lenny Jones
  • Initials: Jane Frances Edwards vs. J. F. Edwards
  • Missing middle names: John Frederick Ferguson vs. John Ferguson
  • Other name elements that are commonly dropped, for instance:
    • Arabic names containing “al” frequently leave it out: Faruq al-Bustani vs. Faruq Bustani
    • Spanish names often drop the second last name: Rafael Nadal Parera vs. Rafael Nadal
  • Simple misspellings and spelling variations, including names that sound alike: Olivia Frank vs. Olivia Franke
  • Name order variations: Kim Soo-hyun vs. Soo-hyun Kim (in Asian names the family name traditionally comes first, but they sometimes exhibit the Western order.)
  • Transliteration variants: Aisha al-Said vs. Ayesha el-Saeed (Given that Arabic is written in a script different from Latin, it has to be transliterated into English, and there is no single transliteration standard. Consequently, differences in English spelling of an Arabic name often occur.)

What makes name matching particularly difficult is that several variations can occur in the same name. For instance: John William Dolittle vs. Jack Doolittle.

For other examples of types of name variation, see here.

A Machine-Learning-Based, AI Approach to Fuzzy Name Matching

The solution to these challenges is a technology called Advanced Fuzzy Name Matching. Advanced Fuzzy Name Matching is an AI technology that uses intelligent machine learning algorithms to automatically learn a very large collection of probabilistic name matching rules from real-world, large-scale name variation data. Since the rules are learned from actual data, they are not constrained by humans’ limited knowledge of possible name matches. They reflect countless name variants that occur in the real world, allowing this approach to produce more accurate matching.

Advanced Fuzzy Name Matching also has automated name ethnicity detection capabilities and applies the most suitable matching models to names based on their ethnicity value. This substantially reduces the number of false positives and false negatives. This approach is much more accurate than such standard algorithms as Soundex and Edit Distance. For more information on name matching algorithms, see here.

Advanced Fuzzy Name Matching also processes names at scale in real time, enabling it to reach the extremely high matching speed required for on-line credit card transactions. It also provides a similarity score that can be used to set thresholds for match, no match, and an in-between gray area match, which may require further risk mitigation strategies.

Summary

Advanced Fuzzy Name Matching is a technology that provides fast and accurate matching of a cardholder’s name on a card against what the customer has provided during the transaction.

It helps protect against credit card fraud, especially for card-not-present transactions.