liveBook · Manning

Named entity recognition

As explained in Machine Learning System Design

Named entity recognition is a natural-language-processing technique that identifies and classifies mentions of entities in text.

Named entity recognition identifies and classifies entity mentions in text. When a deployed model misses customer-important entities and retraining is too slow, a knowledge-base check can provide a hint before inference. section 13.4

Fast correction for missed entities

A team checked incoming text against a knowledge base before running its model. Adding a missing entity took about a minute and could be done by a nontechnical person, fixing the problem without retraining. The technique augments model inputs rather than overriding model outputs. section 13.4

How Machine Learning System Design covers it

How deep it goes
Explains knowledge-base input augmentation and contrasts it with output overrides, without detailing entity-recognition model training.
Examples
Uses a customer-facing system that checks a knowledge base before inference and accepts entries added by nontechnical staff.
What it assumes
Assumes an existing model and knowledge base; does not teach entity recognition or model training from first principles.
What you can do after
Apply knowledge-base hints to correct missed entities without retraining the model.
How it connects
Connects named entity recognition with knowledge bases, model inference, retraining, and overrides.
Cover of Machine Learning System Design
liveBookmanning.com
Read Machine Learning System Design on liveBook Get the book at manning.com

Also explained in

Compare how 4 books explain named entity recognition →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Machine Learning System Design. The section links point to the parts of the book it draws on.