Adding Entity Types for Named Entity Recognition

, ,
1

Is there an expanded database or menu of entity types already setup aside from those mentioned below?

For example (Classification) would be useful for classifying "Division II" sports league from "Division I", but it differs from the Law entity. Is there anything close to this already available?

Thanks
Kevin

Reference from website:

Named Entity Recognition

TIP: UNDERSTANDING ENTITY TYPES

You can also use spacy.explain to get the description for the string representation of an entity label. For example, spacy.explain("LANGUAGE") will return “any named language”.

Models trained on the OntoNotes 5 corpus support the following entity types:

2

The trained NER models only have those entity types. Large numbers of entity types tend to be difficult to annotate for, so I'd probably caution against introducing a very fine-grained scheme. You might do better matching candidates to a knowledge base, and inferring types from there.

You can also add entities to spaCy's models. Sometimes the model really struggles with this though: it's a bit harder to reason about what categories will make it easy to update the model.