AI NER
Introduction
The Lettria API can detect entities in text, and automatically classify and label the text according to the type of entity that it represents, such as a person, location, organization, and so on.
To do this, our algorithms have been trained on large amounts of annotated text data, manually labeled according to the entities it contains.
We use a combination of rule-based and machine learning algorithms, along with custom solutions that you can build on the Lettria platform, to accurately detect entities in text.
Format
Detected entities are received inside an Array() A comprehensive list of entities can be found here.
Example
[
{ "type": "PERSON", "source": "John", "value": null, "index": 0 },
{ "type": "PERSON", "source": "Jean Michel", "value": null, "index": 5 },
{ "type": "ORGANIZATION", "source": "Apple", "value": null, "index": 14 }
]
Entity
Entity Format
| Key | Type | Description |
|---|---|---|
type | String | Entity type |
source | String | Entity source |
value | Number | Entity value |
index | Number | Entity index |
Entity Example
{ "type": "ORGANIZATION", "source": "Apple", "value": null, "index": 14 }