| { |
| "$schema": "https://json-schema.org/draft/2020-12/schema", |
| "title": "Token-level POS, lemma and morphology annotation", |
| "type": "object", |
| "required": ["sentence_id", "language", "tokens"], |
| "additionalProperties": false, |
| "properties": { |
| "sentence_id": { |
| "type": "string" |
| }, |
| "language": { |
| "type": "string" |
| }, |
| "tokens": { |
| "type": "array", |
| "items": { |
| "type": "object", |
| "required": ["surface", "lemma", "upos", "features", "confidence", "comment"], |
| "additionalProperties": false, |
| "properties": { |
| "surface": { |
| "type": "string" |
| }, |
| "lemma": { |
| "type": ["string", "null"] |
| }, |
| "upos": { |
| "type": "string", |
| "enum": [ |
| "ADJ", |
| "ADP", |
| "ADV", |
| "AUX", |
| "CCONJ", |
| "DET", |
| "INTJ", |
| "NOUN", |
| "NUM", |
| "PART", |
| "PRON", |
| "PROPN", |
| "PUNCT", |
| "SCONJ", |
| "SYM", |
| "VERB", |
| "X" |
| ] |
| }, |
| "features": { |
| "type": "object", |
| "additionalProperties": false, |
| "properties": { |
| "Case": { |
| "type": ["string", "null"] |
| }, |
| "Number": { |
| "type": ["string", "null"] |
| }, |
| "Gender": { |
| "type": ["string", "null"] |
| }, |
| "Person": { |
| "type": ["string", "null"] |
| }, |
| "Tense": { |
| "type": ["string", "null"] |
| }, |
| "Mood": { |
| "type": ["string", "null"] |
| }, |
| "Voice": { |
| "type": ["string", "null"] |
| } |
| } |
| }, |
| "confidence": { |
| "type": "string", |
| "enum": ["low", "medium", "high"] |
| }, |
| "comment": { |
| "type": ["string", "null"] |
| } |
| } |
| } |
| } |
| } |
| } |
|
|
