The G3 API is a general purpose NLU API. See this page for a non-technical overview. We also provide specialized Media, Voice of the customer, and Intent API.
Analyses: The API can perform one or more linguistic analyses: language detection, sentiment analysis, entity recognition, semantic tagging, relations.
Supported languages: We support analyses for English, Czech, Dutch, French, German, Polish, Slovak, and Spanish. Some features are not supported for all languages and/or domains. See Domains for more info.
Input preprocessing can extract text from various data formats, remove advertisement, segment text into paragraphs, fix common spelling errors, and (for Czech) add diacritics.
Genres and Domains: except generic analysis, we offer workflows tuned to media articles and customer feedback.
Integration: The API can be accessed with the help of our Python, Kotlin or Typescript SDK, or via Keboola App.
See Swagger documentation for technical details and an interactive demo.
For all calls, supply your API key. Sign up to obtain the key.
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 document larger than 500 KB. However, even analysis of smaller documents might time-out if they contain many entities.
Quick Start – get a simple script analyzing text in a jiffy.
Guide – overview and explanation of basic features of the API.
API Reference – API reference describing individual components of the request and response.