Fuzzy Name Matching Helps to Verify Payee Information to Reduce Misdirected Payment Fraud

What is Misdirected Payment Fraud?

In an era where digital transactions are the norm, misdirected payment fraud (aka authorized push payment fraud) is a growing concern. Misdirected payment fraud is when a payer thinks they are making a payment to their intended payee, but instead they’ve been tricked into sending the payment to a fraudster. For instance, in one common scenario, a fraudster tricks an individual into authorizing a large transfer from their bank account to one under the control of the fraudster. Claiming to be an employee of the individual’s bank, the fraudster tells the individual that he/she has been the victim of a fraud and so has to transfer the money quickly to a different bank account. Doing this is an especially big mistake for those making an instant transfer, since such instant transfers are not reversible once executed.

Reducing Misdirected Payment Fraud with Verification of Payee through Name Matching

But there is a way to reduce this type of fraud. To illustrate how, let’s take a look at what the European Union is doing.

Since its debut in 2017, the European Union’s Single Euro Payments Area (SEPA) Instant Credit Transfer has streamlined the handling of money transfers for financial institutions and payment service providers (PSPs), enabling them to deliver on the promise of real-time, cross-border euro payments throughout thirty-six countries and territories.

A critical piece of SEPA Instant Credit Transfer is the introduction of the Verification of Payee (VoP) scheme to curtail misdirected payment fraud. In this Verification of Payee scheme, just before a payment is to be made, the payer’s PSP instantly sends a request to the payee’s PSP or bank to verify the intended payee’s name along with the IBAN number and potentially an identifier. The payee’s PSP or bank instantly verifies whether the received data matches and provides a response (e.g., match, no match, close match with the name of the payee, match/verification check not possible). The payer’s PSP or bank then immediately passes on the response to the payer, who can then make a decision whether to proceed with the money transfer.

For VoP to be effective, it must be performed accurately and in just a few seconds. An AI technology, Fuzzy Name Matching, can accurately and speedily determiner whether the intended payee’s name matches what the payee’s PSP or bank has on record.

(Needless to say, VoP would also catch accidental misdirections of payments in addition to the fraudulent ones.)

Names Can Exhibit Legitimate Variations

A major challenge in verifying the payee is that names can legitimately vary for a whole host of reasons. In other words, a mismatch between the intended payee name and the name on record at the payee’s PSP or bank should not always result in a failed Verification of Payee. Some variations or mismatches are acceptable, others are not.

For instance, person names can vary in multiple ways:

• • Simple misspellings and spelling variations, including names that sound alike: Jean vs. Gene, Francis vs. Frances
  • Initials: John Frederick Smith vs. J.F. Smith
  • Missing name elements: John Robert Reynolds vs. John Reynolds
  • Nicknames: William vs. Bill
  • Name Order Variants: Park Jae-in vs. Jae-in Park. (Asian names traditionally place the surname, e.g., Park, first, but they sometimes follow the Western order.)
  • Transliteration variants: Tariq Fattah el-Hussain vs. Tareq Fatah al-Hussein (A language like Arabic is written in a script different from Latin and also has some sounds that don’t occur in English. When transforming the name from Arabic letters to English ones, differences in spelling frequently arise.)

What makes name matching particularly difficult is that several variations can occur in the same name (as illustrated in the last example above).

Organization names can also vary a great deal. Here are some examples of the ways in which business names can vary:

• • Full form vs. acronym: British Broadcasting Corporation vs. BBC
  • Full form vs. abbreviated form: Federal Express vs. FedEx
  • Presence vs. absence of corporate designators: BlueHill Corp. vs. BlueHill

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

Matching Names Across Different Writing Systems Is Particularly Challenging

It’s bad enough if the names only come in Latin script, but they may also come in a non-Latin script that has to be verified against the expected Latin version.

• • For example, the money transfer may state the payee’s name is Kim Jong-Park, but the name on the destination account may be 김종박.
  • Or a payment for Alexey Rokossovsky may be met by Алексей Рокоссовский
  • Or Tariq al-Bustani طارق البستاني

In all these cases, the names refer to the same person but are in different writing systems.

For more information on the challenges of cross-language name matching, see here.

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

The solution to the challenges posed by Verification of Payee is Advanced Fuzzy Name Matching. Advanced Fuzzy Name Matching uses intelligent machine learning algorithms to automatically learn a collection of probabilistic name matching rules from large-scale, real-world name variant data. Since the rules are learned from real data, they are not limited by humans’ knowledge of what are possible name matches. They reflect countless name variants that occur in the real world.

In the case of cross-language name matching, the matching is performed directly, with no need to translate non-Latin scripts into Latin prior to matching. This avoids the problem of mistakes introduced by the translation process. Matching directly between the names in different scripts achieves higher accuracy.

Advanced Fuzzy Name Matching provides a similarity score that can be used to set thresholds for match, no match, and an in-between weaker match that can be presented to the payer for confirmation before proceeding with the payment.

Summary

Advanced Fuzzy Name Matching allows financial institutions making money transfers to send the payee’s information for verification by the payee’s PSP or bank and instantly receive back a reply as to whether the names match. Advanced Fuzzy Name Matching enables Verification of Payee in instant transfers and provides payers with the confidence that their money is going to the intended payee.