Skip to content
BENIBIX Learn

Named Entity Recognition (NER)

AI🔤 NLP & LanguageLanguage Tasks

Also known as NER, Entity Extraction

Finding names of people, places, dates and firms in text.

Think of it like

Think of it like you read a story with a pen. You ring every name, place and date.

Example

Jack's firm gets 500 contracts a week. NER pulls out each buyer, seller, date and price and puts them into a table.

Why it matters

Most useful facts in text hang on names. Finding them is the first step to turning text into data.

Definition

Named entity recognition finds the names of things in a text and says what kind each one is. The usual kinds are people, places, groups, dates and sums of money. Take 'Lucy flew to Paris on Monday'. It marks Lucy as a person, Paris as a place and Monday as a date. It turns loose text into neat records that a program can store and search.
Each name in the sentence is marked with its kind: a person, a place and a date.

Common mistake

Expecting it to know which Paris is meant. It can mark 'Paris' as a place, but the town in Texas and the capital of France look the same.

Key takeaways

  1. 1
    Finds names. In plain text.
  2. 2
    Gives each a kind. Person, place, date, group.
  3. 3
    Text to records. Ready to store and search.

Compare