What's in a Name? Automated Indexing

What’s in a Name?

A grantor by any other name would be… inaccurate…


When it comes to document indexing or any kind of data capture, it is very common to need to find the names of various parties or document contributors. This can be a grantor or grantee in a Grant Deed or a patient in an external lab result. One of the advantages to having a person manually review documents is that they can identify the type of information that they are looking at. When a person sees two names separated by a space, the user knows that the first word is most likely the first name and the second word is most likely the second name. If the subsequent document is the same but there are two names separated by a comma, that person knows that now the first word is likely the last name. That person could also look at 99 Grant Deeds and pick out the individual grantee as a person on all of them and be able to recognize that on the hundredth, the grantee is the six-word name of a trust instead of a person. An ideal automated solution would have the same ability to recognize these differences and understand that something like a name can mean different things and be presented in different ways.


Name is a field that highlights the distinction between just capturing data from a document and intelligently capturing data from a document. Using the grantor on a Grant Deed as an example, the name could be a company, a bank, an individual, or any combination or multiples of those. One of the qualities that sets Extract apart is our ability to recognize different types of parties on the same type of document. There are two pieces to this puzzle.

  1. Correctly identifying the data on the document.
  2. Placing that in a downstream system in the desired format.

An intelligent solution can recognize the grantor on one document as National Bank of Verona and index that as “National Bank of Verona” and recognize the grantor on another document as William Shakespeare and index that as “Shakespeare, William.” This intelligence and flexibility is something that we excel at.


Even within the names of individuals, there can be a lot of variation in how they are formatted. There is an increasing number of apostrophes and dashes in first and last names. When a healthcare organization receives a faxed lab report, it is important that they are correctly identifying the patient that those results belong to. If a paper document lists a patient as “Montague-Capulet, Romeo,” an intelligent solution can identify the patient’s first name as “Romeo” and last name as “Montague-Capulet” and understand that this is different from other patients whose names might be “Romeo Montague,” “Romeo Capulet,” or even “Romeo Capulet-Montague.” This is a strength of Extract and enables us to correctly attach lab results to the appropriate patients in EMRs.


If parting is such sweet sorrow, then the only thing sadder is not correctly capturing names from your scanned documents. Using Extract’s data capture solution can eliminate this concern and ensure that all of the information that you get from your documents is accurate and formatted the way that you expect. To learn more about automated document classification and indexing, check out the next upcoming Monthly Webinar where we will be demonstrating the Extract solution in-action!


Monthly Software Demonstration

Monday, January 30th at 1pm CST

Sit back and relax as we take you through the Extract Systems™ Platform and the many features and capabilities Extract offers. This is the perfect opportunity to ask any questions you may have and explore the numerous benefits the Extract Systems™ Platform can offer your organization.

About the Author: Eric Davidson

Eric Davidson is a Customer Support Specialist & Project Manager at Extract. He is heavily involved in the design and development of Extract’s data entry interface. Prior to Extract, Eric spent 3 and a half years working in Healthcare IT and learning the ins and outs of EMRs at Epic.