Amazon Comprehend · Schema

Amazon Comprehend Entity

Schema for an Amazon Comprehend entity detection result, representing named entities identified in text through NLP analysis.

Machine LearningNatural Language ProcessingNLPText Analysis

Properties

Name Type Description
Entities array A collection of entities identified in the input text.
Sentiment string The inferred sentiment detected in the input text.
SentimentScore object The confidence scores for each sentiment type.
KeyPhrases array A collection of key phrases identified in the input text.
Languages array The languages detected in the input text.
View JSON Schema on GitHub

JSON Schema

amazon-comprehend-entity-schema.json Raw ↑
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://example.com/schemas/amazon-comprehend-entity.json",
  "title": "Amazon Comprehend Entity",
  "description": "Schema for an Amazon Comprehend entity detection result, representing named entities identified in text through NLP analysis.",
  "type": "object",
  "properties": {
    "Entities": {
      "type": "array",
      "description": "A collection of entities identified in the input text.",
      "items": {
        "$ref": "#/$defs/Entity"
      }
    },
    "Sentiment": {
      "type": "string",
      "description": "The inferred sentiment detected in the input text.",
      "enum": ["POSITIVE", "NEGATIVE", "NEUTRAL", "MIXED"]
    },
    "SentimentScore": {
      "type": "object",
      "description": "The confidence scores for each sentiment type.",
      "properties": {
        "Positive": {
          "type": "number",
          "minimum": 0,
          "maximum": 1
        },
        "Negative": {
          "type": "number",
          "minimum": 0,
          "maximum": 1
        },
        "Neutral": {
          "type": "number",
          "minimum": 0,
          "maximum": 1
        },
        "Mixed": {
          "type": "number",
          "minimum": 0,
          "maximum": 1
        }
      }
    },
    "KeyPhrases": {
      "type": "array",
      "description": "A collection of key phrases identified in the input text.",
      "items": {
        "$ref": "#/$defs/KeyPhrase"
      }
    },
    "Languages": {
      "type": "array",
      "description": "The languages detected in the input text.",
      "items": {
        "$ref": "#/$defs/DominantLanguage"
      }
    }
  },
  "$defs": {
    "Entity": {
      "type": "object",
      "description": "A named entity identified in the text.",
      "properties": {
        "Score": {
          "type": "number",
          "description": "The level of confidence that Comprehend has in the accuracy of the detection.",
          "minimum": 0,
          "maximum": 1
        },
        "Type": {
          "type": "string",
          "description": "The entity type.",
          "enum": [
            "PERSON",
            "LOCATION",
            "ORGANIZATION",
            "COMMERCIAL_ITEM",
            "EVENT",
            "DATE",
            "QUANTITY",
            "TITLE",
            "OTHER"
          ]
        },
        "Text": {
          "type": "string",
          "description": "The text of the entity."
        },
        "BeginOffset": {
          "type": "integer",
          "description": "The zero-based offset in the input text where the entity begins.",
          "minimum": 0
        },
        "EndOffset": {
          "type": "integer",
          "description": "The zero-based offset in the input text after the entity ends.",
          "minimum": 0
        }
      }
    },
    "KeyPhrase": {
      "type": "object",
      "description": "A key phrase detected in the text.",
      "properties": {
        "Score": {
          "type": "number",
          "minimum": 0,
          "maximum": 1
        },
        "Text": {
          "type": "string",
          "description": "The text of the key phrase."
        },
        "BeginOffset": {
          "type": "integer",
          "minimum": 0
        },
        "EndOffset": {
          "type": "integer",
          "minimum": 0
        }
      }
    },
    "DominantLanguage": {
      "type": "object",
      "description": "A language detected in the text.",
      "properties": {
        "LanguageCode": {
          "type": "string",
          "description": "The RFC 5646 language code for the dominant language."
        },
        "Score": {
          "type": "number",
          "description": "The confidence score for the detected language.",
          "minimum": 0,
          "maximum": 1
        }
      }
    }
  }
}