Skip to content

Document Comprehension


The API output is divided into three levels, with the Document level being the highest. The Document Object contains metadata about the analyzed text, as well as subtypes for further analysis such as Sentence, Coreference, Emotion, and Patterns.

The Sentence subtype breaks down the text into individual sentences and provides information about each one, while the Coreference subtype identifies and groups together mentions of the same entity. The Emotion subtype detects and categorizes emotions expressed in the text, and the Patterns subtype identifies common patterns or themes.

The Document returned by the API offers a comprehensive analysis of the text, providing valuable insights into its content and meaning.


Documents are received inside an Array()

languageStringLanguage of the document
source_pureStringSource of the document without normalisation
sentencesSentence [ ]Sentences inside the document
coreferenceCoreferenceCorefence inside the document
sentimentfloatSentiment value at document level
emotionsEmotion [ ]Sentences inside the document
patternsMatched PatternsSentences inside the document

Next steps