3.3 Extracting information : Named Entity Recognition (NER)

3.3.1 What are named entities ?

Named entities are generally proper nouns that refer to specific entities, such as a person, organization, location, date, etc. If we consider this sentence as an example: Mount Everest is the tallest mountain above sea level, NER should detect Mount Everest as a named entity of type location as it refers to a specific entity.
Some other examples of named entities are listed in the following table:

Named Entity
ORGANIZATION United Nations Organization, UNICEF, Microsoft
PERSON Novak Djokovic, Beyoncé, Scarlett Johansson
LOCATION Mount Everest, River Nile, Machu Picchu Archaeological Park
DATE 3rd April 1988, 7 June
TIME 8:45 A.M., one-thirty am
GPE France, Liechtenstein, Democratic Republic of Congo
MONEY 7 million dollars, 73.01 INR

What should be considered a named entity (NE) in a text is open for discussion and depends on the information one wants to extract. However, the set of named entity classes that are widely used contains the three fundamental entity types, the person (PER), location (LOC), and organization (ORG), collectively referred to as the enamex since the MUC-6 competition (Grishman et al. 1996).

In the following sentence, try to identify the named entities :