NEW
Free Consultation Today
x

Give our baseline algorithm a try and see how modern NLP models perform similarity matching between personal names.

Our NLP algorithms learn a deeper relationship between records than what is possible with traditional fuzzy matching, rules, and distance algorithms.

Those legacy algorithms are proven to struggle to scale as you ramp up your NLP pipeline to more use cases.

Give Our API a Try

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Name Similarity Score

Why are deep learning language models more accurate?


These large language models focus on learning a deeper relationship between the compared fields than what can be learned strictly based on semantics, rules, or fuzzy matching.

Through fine-tuning and in-domain training we show these models exactly how to solve both common and edge cases that these older methods struggle to cover.

With all that said, the key benefit these models show is how they scale their data variance coverage as your data scales over time.


We take this baseline model and fine-tune it for your specific use case to explode the in-domain accuracy.

Looking to use this API? Let's chat!

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

What can I use this API for?


- Automated customer record processing

- Data deduplication

- Record matching

- Smart login systems

- Document processing

Most Common Challenges of Name Matching


- Missing Spaces & Hyphens: MattPayne = Matt Payne = Matt--+Payne

- Nicknames: Jake = Jacob or Matt = Matthew

- Initials: J.J. Wills = James Johnson Wills

- Missing Parts of the Name: Mary Nicole Jackson = Mary Jackson = Mary N. Jackson

- Different spelling same name (This tricks those fuzzy matchers!): Stuart = Stewart

Let's solve your business problem

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.