Comprehend is Amazon’s natural language processing tool, announced last year to help businesses selectively extract words and phrases. Amazon is now adding to its capabilities and has announced a brand new feature that will let developers build lists of specific words and phrases without needing to possess domain specific knowledge in machine learning.
The announcement comes barely a week before the commencement of Amazon’s upcoming Re:invent conference.
Speaking about the development, Matt Wood, GM for deep learning and AI, said:
Today we are excited to bring new customization features to Comprehend, which allow developers to extend Comprehend to identify natural language terms and classify text which is specialized to their team, business or industry
So basically, Amazon will be doing the heavy lifting while developers can go on adding customized lists without having to first wet their feet with either machine learning ot natural language processing.
Under the hood, Comprehend will do the heavy lifting to build, train, and host the customized machine learning models, and make those models available through a private API.
So basically, the system will function as follows: Developers will need to define a list of custom entities. Examples could include medical or legal jargon. Next up, they will have to expose a list of these entities to Comprehend. Amazon will then automatically learn to identify the custom language and it will go about building a customized model that responds to the entities in the list.
Once you add in some logical lists, Comprehend will start to categorize the entities and move them through a workflow to the appropriate personnel, or store them to be used later by an application.
Through as few as 50 examples, Comprehend will automatically train a custom classification model that can be used to categorize all your documents. You could group support emails by department, social media posts by product, or analyst reports by business unit.
The feature is ready for consumption, starting today.
A bibliophile and a business enthusiast.