Natural Language Processing for Legal Documents

Our Sifters

Built on the experience of tens of thousands of people, each Sifter® is trained to read text and look for a specific concept. It reads novel language, learns from experience, and improves over time.

A Sifter uses AI, specifically natural language processing and machine learning.

Artificial Intelligence and my brown dog

The dog is brown.
They call him Charlie.
Charlie is a Labrador retriever.
The brown puppy is named Charlie.

People use different words to describe the same thing, like Charlie the dog. The variety makes language interesting ... and legal documents painful to read.

Each Sifter uses two types of artificial intelligence to solve this problem: machine learning (ML) and natural language processing (NLP).

NLP teaches computers to interpret human language as written, regardless of style. Each Sifter uses NLP to convert text into meaning.

We train our Sifters with machine learning to identify a specific concept like a consequential damages waiver. Our lawyers find hundreds or thousands of examples of a specific concept. Our Data Science team uses machine learning to find the intangible thread that binds the writing of lots of lawyers, all with their own grammatical flair.

As more of our clients use a Sifter, it improves. Our Sifter Trainer, for example, retrains Sifters within 5-10 days of receiving client feedback.