Skip to main content

NlpRequest

NLP request for Media Assistant v2, based on the General API request.

See the semantic tagging guide for an example of using an NLP request to obtain tags.

The API is intended for analysis of typical newspaper articles, not books.
Requests larger than 500 KB are rejected.
Even smaller documents may time out if they contain many entities.
If your use case involves large articles, please contact us.

Fields

PropertyRequiredTypeDescription
idTruestringUnique identifier of the article
titleFalsestringTitle or subject of the article; mutually exclusive with paraSpecs
leadFalsestringLead or abstract of the article; mutually exclusive with paraSpecs
textFalsestringMain text or body of the article; mutually exclusive with paraSpecs
paraSpecsFalseList[ParaSpec]Paragraph specifications; mutually exclusive with title, lead, and text
domicileFalsestringLocation part of the dateline indicating the origin, e.g., Paris, France
analysesFalseList[string][option] Analyses to return; default: all
languageFalsestring

[option] ISO 639-1 language code of the article; auto-detection is used if omitted. See supported languages.

langDetectPriorFalsestring

[option] Language detection prior, e.g., de,en. See supported priors.

presentationLanguageFalsestring

[option] Presentation language of the results in ISO 639-1 format, e.g., fr. See details.

categoriesFalseList[Category][option] Categories (and entities) of the article
referenceDateFalsestring

[option] Reference date/time for the analysis; accepts:

  • NOW – current date and time
  • yyyy-mm-dd – date only, GMT timezone, e.g., 2024-06-21
  • ISO date-time – as accepted by java.time.format.DateTimeFormatter.ISO_DATE_TIME's parse method, e.g., 2024-06-21T10:15:30 or 2024-06-21T10:15:30+01:00
  • Any other or invalid value is treated as NOW.
returnMentionsFalseboolean

[option] Whether to return entity/tag/relation mentions; default: false. See the mentions guide.

returnItemSentimentFalseboolean[option] Whether to return sentiment for entities, mentions, tags, relations, etc.; default: false
returnGkbPropertiesFalseboolean[option] Whether to return GKB properties for entities/tags; default: false
metadataFalseMap[string, string]Additional metadata as key-value pairs, e.g., {"author": "John Smith", "url": "https://example.com/abc" }
recommendationFalseRecommendationRecommendation info for the article.