Entity Detection

Task Description

The Entity Detection endpoint identifies and extracts both traditional named entities and novel entity types from text. It goes beyond standard entity recognition by detecting abstract concepts, domain-specific entities, and creating meaningful entity categorizations with descriptive explanations.

Inputs

• Text: The transcript or text content to analyze for entities.

Output

Description

The API returns a JSON object containing an array of entity objects. Each entity object includes:

• text: The exact text span identified as an entity
• type: A meaningful classification or concept for the entity
• description: An explanation of what this entity type represents

Example

1
{
2
  "entities": [
3
    {
4
      "text": "Apple Inc.",
5
      "type": "Organization",
6
      "description": "A company or formal institution"
7
    },
8
    {
9
      "text": "machine learning",
10
      "type": "Technical Concept",
11
      "description": "A field of artificial intelligence focused on systems that can learn from data"
12
    },
13
    {
14
      "text": "climate crisis",
15
      "type": "Environmental Issue",
16
      "description": "A significant challenge related to changes in global climate patterns"
17
    }
18
  ]
19
}