Answers: TextSearch for Names - how to support Nicknames and Longer Names?

Search
1

Hi Team -

I have providers for my healthcare company in Answers and I have made First Name and Last name searchable with textSearch. I didnt use NLP because there is some overlap between First name and Last name so I needed the net to be wider.

However, the issue I have is as follows: I have some doctors who go by nicknames and I want to be able to search for part of their name (aka first few letters) and have it surface (for example, Alex for Alexander or Ken for Kenneth). Other times, doctors have very long last names that are hard to spell and I want to be able to just type in the first 4 letter and have that doctor surface.

What would you recommend here? Adding a keyword for each one is a bit cumbersome because have 2000 doctors. Is there a wider solution or is this possible to support?

Thanks!
Alyssa

2

Could you create custom fields and export to Excel and use the LEFT() function?

i.e. for the nicknames, create c_nickname and then export all doctors names, use the left function to get the 1st 3 letters of the name, and then re-upload to the nickname field.

Also, not sure if it’s possible to set up a query rule to use regex somehow, since the query might not be exactly the 3 letters (from the example above) - with the action being a filter to set the results set based on the nickname/name field.

3

Thanks for the ideas Ariana! I’ll try query rules to start and then perhaps the left function after that makes sense though I would worry about nicknames that are 3 vs 4 letters etc.

4

Of course! The best case scenario would be the client providing the nicknames and having a c_nickname as a searchable field (I think I’ve seen this in some healthcare accounts).

5

Hey Alyssa,

There are a few things that could help here:

  1. We have a feature on the roadmap that will allow you to use Knowledge Graph data to populate query suggestions, so that if you type the first couple letters of a doctor’s name their full name will pop up in autocomplete.
  2. We have yet another feature that will incorporate data from the Knowledge Graph into our spell check model, so that if you misspell someone’s last name, we will (often) catch it in spell-check and suggest the correct spelling.
  3. For the Alex/Alexander issue, I think the best way to solve this problem will be to add use a Nickname field with the alternative name and to make that searchable. Presently, we have no way to resolve the nicknames to their expanded names.

Finally, I would note that we do use typo tolerance, so assuming that the user’s spelling is “close enough” to the doctor’s name, it should still show up.

Does this help?