Name Matching Is Critical for Accurate and Effective Employment Background Checks

Background Checks Are Essential in Many Business Areas

Background checks are extremely common nowadays with an estimated 95% of US businesses conducting some form of pre-employment screening.

Background checks are critical, among other things, for ensuring workplace safety and protecting a company’s reputation and assets. The aim is to uncover any red flags such as criminal history, fraud, or falsified credentials.

Many different types of organizations need to vet their prospective employees:

• • Businesses
  • Government agencies
  • Schools
  • Non-profits (employees, members, volunteers)
  • Staffing firms
  • etc.

Background checks are often performed by specialized companies. In fact, this is a fast growing industry projected to reach $39-$40 billion by the early 2032 driven by the growing need for compliance with legal processes and rising awareness of the importance of trust and security in hiring.

Name Matching, an AI Machine-Learning technology, can expedite the background check process and free up human teams from what can be a tedious process.

Background Checks Are a Challenging Task

Background checks are a challenging task. Not only are unique identifiers such as Social Security Numbers rarely provided in court documents, but there is no national database of criminal records. Instead, criminal records are maintained by each of the 3,144 counties in the U.S. with no consistency in the personal identifiers provided within records. One jurisdiction may have full name, date of birth (DoB), and address. Other jurisdictions may only include full name and age. All these variations make it hard to identify every record that belongs to a person.

To compound the problem, each data field can exhibit one or more types of variations. For instance:

• • Personal Name
    • Many married women take their husband’s surname.
    • A legal name may change for many reasons including marriage, divorce, etc.
    • Nicknames: Joseph vs. Joe
    • Spelling variations: Karen vs. Karyn vs. Karin; Sean vs. Shawn vs. Shaun
    • Names may miss one or more name components such as middle names: Francis Henderson vs. Francis Anthony Henderson
    • Word order differences: John Robinson vs. Robinson, John
    • The use of initials: Frederik John Henderson vs. Frederik J. Henderson
    • The use of nicknames: Edward James vs. Ted/Teddy James
    • Apostrophes are frequently omitted: O’Connor vs. OConnor
    • Simple typos, which are more common with less familiar names and spellings
    • Cultural naming conventions, such as:
      • Spanish surnames, which include a patronym and a matronym, may drop the latter: Raúl Gonzaga Juárez vs. Raúl Gonzaga
      • Arabic names which contain the definite article “al” frequently leave it out: Said al-Bustani vs. Said Bustani

• • Date of Birth
    • 10/9/1965 1965 (some jurisdictions report only the year)
    • October 9, 1965 Oct. 9, 1965
    • October 9, 1965 9 October, 1965

• • Place of Birth
    • Winchester, Mass. Winchester, Massachusetts vs. Winchester, MA

• • Address:
    • One Beaver Lane vs. 1 Beaver Lane
    • 39 Dedham Street vs. 39 Dedham St.
    • 44 Anderson Place, New York vs. 44 Anderson Pl., NY (This is an example of more than one type of variation).

Name Matching Supports Accurate and Fast Background Checks

Fortunately, an AI technology is available that eases the challenges of matching immensely: Name Matching.

Name Matching handles the above variations and also many others. It provides the required fuzzy name matching using Machine Learning techniques to create a different trained matching model specific to each entity type.

Each model is trained to match the likely variations in each entity type (e.g., people, organizations, dates, places, addresses). Each search query field (e.g., name, date of birth, place of birth) is matched separately against its counterpart, and a similarity score is generated. These field scores are combined into a single overall score for the entire record. Business logic rules can set thresholds for how high an overall score as well as each field score needs to be for the candidate historical record to be returned as a potential match.

For example, here’s a sample user query:

Name:                         DoB:                  PoB:
Sean Andersen                 10/9/1976             Woburn, MA

And here is a set of records that may be returned, ranked in decreasing order of similarity:

Name:                         DoB:                  PoB:
Sean M. Andersen              10/09/1976            Woburn, MA
Shawn Andreeson               1976                  Woburn, Mass.
Sean Anderson                 12/4/1978             Waburn, MA
John Andersen                 10/09/1976            Woburn, MA

Summary

Background checks are a necessary step in onboarding prospective employees and monitoring current ones. Name Matching is a fast, cost-effective, and accurate solution that expedites the background check process. It offers powerful AI matching algorithms tailored to the requirements of different types of names and is able to rank candidate matches based on all the data fields in a record. And it does this at scale, enabling rapid background checks.