What does “Named Entity Recognition (NER)” mean?
An NLP task that identifies entities such as people, places, dates, or organizations.
How Named Entity Recognition (NER) Works
Text: 'Ada Lovelace visited London in 1843.'
NER model
Ada Lovelace = PERSON
London = PLACE, 1843 = DATE.
Practical Example
A news-analysis tool scans articles and highlights each person, company, and place mentioned, so a reader can see who and what a story involves without reading every line.
Why It Matters
NER turns messy text into structured facts that can be searched and counted. It is a building block for search, compliance, and data extraction.
Common Questions About Named Entity Recognition (NER)
Q: Which entity types does NER find?
Common ones are people, organizations, locations, dates, and amounts; custom types such as drug names can be added.
Q: Why is it hard?
Names are ambiguous, for example 'Amazon' can be a company or a river, so context is needed.
Q: Do LLMs replace NER models?
They can do it from a prompt, but smaller dedicated NER models are still used for speed and cost.