How Name Matching is Crucial for Transaction Screening as Part of AML

Transaction Screening is a Critical Part of Anti-Money Laundering

AML rules generally require organizations to take serious and prudent steps in detecting and reporting suspicious activity such as money laundering and terrorist financing activity. Examples of AML legislation include the U.S. Patriot Act and the Bank Secrecy Act, both of which have been powerful motivations for organizations to institute strong AML procedures.

Transaction screening is a critical piece of an organization’s overall AML process. It involves screening names of the sender and the beneficiary involved in the transaction against various sanctions lists such as OFAC, UN, EU, and HMT, watch lists, or other specific criteria before it is approved. The aim is to ensure that a restricted party such as a known terrorist or a Politically Exposed Person (PEP) who may have slipped into criminality is not involved in the transaction.

A transaction screening process also has to be able to handle a very large volume of transactions in real time so as to not delay legitimate transactions.

Names Frequently Exhibit Variations

A major challenge in screening the names of the sender and the beneficiary is that names can legitimately vary for many reasons.

For instance, person names can vary in multiple ways:

• • Simple misspellings and spelling variations, including names that sound alike:
    • John Smith vs. John Smyth
    • Sean Moore vs. Shawn Moore
  • Initials:
    • John Edward McClellan vs. J. E. McClellan
  • Missing name elements:
    • Patricia Maureen McMahon vs. Patricia McMahon
    • Abdullah al-Sistani vs. Abdullah Sistani (“al,” the Arabic definite article, is frequently used in names but is also frequently left out)
  • Nicknames:
    • Edward Cole vs. Ed Cole vs. Eddy Cole vs. Eddie Cole
  • Name Order Variants:
    • Lee Ji-eun vs. Ji-eun Lee (Asian names traditionally have the surname come first, but they occasionally exhibit the Western order.)
  • Transliteration variants:
    • Umar Abdullah el-Gharbi vs. Omar Abdallah al-Gharbi (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 frequently occur.)

What makes name matching particularly difficult is that several variations can occur in the same name. For instance:

• • • Joseph Patrick McLaughlin vs. Joe MacLaughlin.

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:
    • Systems Research and Applications Corporation vs. SRA
  • Full form vs. abbreviated form:
    • Federal Express vs. FedEx
  • Presence vs. absence of corporate designators:
    • Smith and Lombaugh LLC vs. Smith and Lombaugh

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

Matching Names Across Different Writing Systems Is Challenging

Sanction lists typically come in Latin script. In a global economy however, names may also come in a non-Latin script that has to be matched against the expected Latin version.

Imagine a financial institution in the Middle East that needs to process a transaction and the names of the sender and beneficiary are in Arabic script and need to be matched against a Latin-based database of individuals:

• • Musa Hussein vs. موسى حسين

This situation occurs with other languages:

• • Wang Yi vs.  王毅 (Chinese)
  • Anastasia Golubeva vs. Анастасия Голубева (Russian)
  • Kim Yong-hyun vs. 김용현 (Korean)
  • Ippei Mizuhara vs. 水原一平 (Japanese)

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 AML Transaction Screening is 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 large-scale, real-world name variant data.

Since the rules are learned from real data, they are not limited by humans’ knowledge of possible name matches. They reflect countless name variants that occur in the real world, allowing this approach to produce more accurate, high recall and high precision matching.

Advanced Fuzzy Name Matching can also handle very high volumes in real time, enabling it to reach the matching speed required by Transaction Screening.

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 also provides a similarity score that can be used to set thresholds for match, no match, and an in-between weaker match that can be evaluated for confirmation before approving the transaction.

Summary

Advanced Fuzzy Name Matching is a technology that provides fast and accurate matching of both the sender and the beneficiary of a transaction against the various databases of restricted entities. Advanced Fuzzy Name Matching allows organizations to perform Transaction Screening in real time in high-volume environments.