Docs / Schemas / classification

Classification 0.8.0

Represent the result of a classification. Supports predicted labels, confidence scores, and vocabulary.

$id

https://oatschema.org/0.8.0/classification

Latest

https://oatschema.org/classification

Versions

Source

v0.8.0 on GitHub

Schema

classification.json Raw

{    "x-license": {        "id": "Apache-2.0",        "notice": "Copyright 2025-2026 The OATS Authors",        "url": "https://github.com/oatschema/oats/blob/main/LICENSE"    },    "$schema": "https://json-schema.org/draft/2020-12/schema",    "$id": "https://oatschema.org/0.8.0/classification",    "title": "Classification",    "x-version": "0.8.0",    "description": "Represent the result of a classification. Supports predicted labels, confidence scores, and vocabulary.",    "type": "object",    "properties": {        "target": {            "type": "string",            "description": "The name of or a reference to the variable being classified (e.g., 'sentiment', 'topic').",            "maxLength": 256        },        "producer": {            "type": "string",            "description": "The creator (model, tool or author) of this classification.",            "maxLength": 1024        },        "score_explanation": {            "type": "string",            "description": "A brief explanation of what the score represents (e.g., 'Model confidence probability [0, 1]', 'Sentiment score [-1, 1]').",            "maxLength": 256        },        "vocabulary": {            "type": "array",            "description": "A list of the possible labels in this classification scheme. For more than 100 labels, use 'vocabulary_url'.",            "maxItems": 100,            "uniqueItems": true,            "items": {                "type": "string",                "description": "A single valid label from the vocabulary.",                "maxLength": 128            }        },        "vocabulary_url": {            "type": "string",            "description": "A URL pointing to a more detailed external vocabulary or ontology.",            "format": "uri",            "maxLength": 2048        },        "labels": {            "type": "array",            "description": "The labels assigned. Can be empty if no labels match.",            "maxItems": 10000,            "minItems": 0,            "items": {                "type": "object",                "properties": {                    "label": {                        "type": "string",                        "description": "The predicted label. Must be a member of 'vocabulary' when it is present.",                        "maxLength": 128                    },                    "score": {                        "type": "number",                        "description": "A normalized score for this label (-1.0 to 1.0). Depending on the use case, this may represent a probability (0 to 1) or a bipolar metric like sentiment (-1 to 1).",                        "minimum": -1,                        "maximum": 1                    },                    "timestamp": {                        "type": "string",                        "format": "date-time",                        "maxLength": 128,                        "description": "The specific time this label applies to. Used for discrete time-series or state logging."                    },                    "attributes": {                        "type": "object",                        "description": "Additional metadata about this specific label.",                        "maxProperties": 16,                        "propertyNames": {                            "maxLength": 128                        },                        "additionalProperties": {                            "anyOf": [                                {                                    "type": "string",                                    "maxLength": 1024                                },                                {                                    "type": "number"                                },                                {                                    "type": "boolean"                                },                                {                                    "type": "null"                                }                            ]                        }                    }                },                "required": [                    "label"                ],                "additionalProperties": false            }        },        "attributes": {            "type": "object",            "description": "Arbitrary metadata relevant to this classification.",            "maxProperties": 16,            "propertyNames": {                "maxLength": 128            },            "additionalProperties": {                "anyOf": [                    {                        "type": "string",                        "maxLength": 1024                    },                    {                        "type": "number"                    },                    {                        "type": "boolean"                    },                    {                        "type": "null"                    }                ]            }        }    },    "required": [        "labels"    ],    "additionalProperties": false}

Examples

Records that pass validation against this schema.

classification-1

classification-1.json

{  "score_explanation": "Model confidence probability, where 1.0 is 100% confident.",  "vocabulary": [      "animal",      "cat",      "dog",      "house",      "vehicle"  ],  "labels": [    {      "label": "animal",      "score": 0.98    },    {      "label": "cat",      "score": 0.95    },    {      "label": "house",      "score": 0.75    }  ]}

Fields

target string

The name of or a reference to the variable being classified (e.g., 'sentiment', 'topic').
- at most 256 characters
producer string

The creator (model, tool or author) of this classification.
- at most 1024 characters
score_explanation string

A brief explanation of what the score represents (e.g., 'Model confidence probability [0, 1]', 'Sentiment score [-1, 1]').
- at most 256 characters
vocabulary array of string

A list of the possible labels in this classification scheme. For more than 100 labels, use 'vocabulary_url'.
Each item: A single valid label from the vocabulary.
- at most 100 items
- items must be unique
- each at most 128 characters
vocabulary_url string

A URL pointing to a more detailed external vocabulary or ontology.
- at most 2048 characters
- format uri
labels array of object required

The labels assigned. Can be empty if no labels match.
- 0 to 10000 items
- label string required
  
  The predicted label. Must be a member of 'vocabulary' when it is present.
  - at most 128 characters
- score number
  
  A normalized score for this label (-1.0 to 1.0). Depending on the use case, this may represent a probability (0 to 1) or a bipolar metric like sentiment (-1 to 1).
  - -1 to 1
- timestamp string
  
  The specific time this label applies to. Used for discrete time-series or state logging.
  - at most 128 characters
  - format date-time
- attributes object
  
  Additional metadata about this specific label.
  - at most 16 properties
  - keys at most 128 characters
  - * string | number | boolean | null
    
    Any other key.
    - string: at most 1024 characters
attributes object

Arbitrary metadata relevant to this classification.
- at most 16 properties
- keys at most 128 characters
- * string | number | boolean | null
  
  Any other key.
  - string: at most 1024 characters