Request

Request for NLP API version 3.

The API is intended for analysis of small and medium sized documents: reviews, newspapers articles, etc., not books. Large documents must be split into smaller units. The API will reject any request larger than 500 KB. However, even analysis of smaller documents might time-out if they contain many entities.

Fields

Property

Required

Type

Description

id

False

string

Unique identifier of the document

title

False

string

The title or subject of the document, when available; mutually exclusive with the paraSpecs parameter. Typically just one paragraph, but may still be split into paragraphs as specified in here.

text

False

string

The main text of the document; mutually exclusive with the paraSpecs parameter. It is split into paragraphs, see here.

paraSpecs

False

List[ParaSpec]

The document paragraphs; mutually exclusive with title and text parameters

analyses

False

List[string]

[option] What analysis to return. Default: all. The options are: all, entities, tags, relations, sentiment, language.

htmlExtractor

False

string

[option] Text extractor to be used when analyzing HTML document

language

False

string

[option] The language of the document as ISO 639-1; auto-detection will be used if omitted. The options are: en, cs, de, es, fr, nl, pl, sk.

langDetectPrior

False

string

[option] The language detection prior; e.g. de,en. Supported priors here.

domain

False

string

[option] The source domain from which the document originates. Supported domains here.

textType

False

string

[option] The type or genre of text; e.g. article,casual,customer-feedback,email,social

referenceDate

False

string

[option] Date to be used for the analysis as a reference; values: “NOW” or in format YYYY-MM-DD

diacritization

False

string

[option] Determines whether to perform text diacritization

returnMentions

False

boolean

[option] Should entity/tag/relation mentions be returned? Default: false

returnItemSentiment

False

boolean

[option] Should entity/mention/tag/relation etc. sentiment be returned? Default: false