This documentation is for Idyl E3 2.0.0 and the most up to date version of this documentation can be found here.

Introduction to Idyl E3

Thank you for using Idyl E3! Idyl E3 is a product of Mountain Fog, Inc.

Idyl E3 (Entity Extraction Engine) performs named-entity extraction from natural language text. For example, given the input text “George Washington was president of the United States.”, Idyl E3 with an English-language person and place models would identify George Washington as a person entity and United States as a place entity.

Idyl E3 is available in multiple editions. To compare the features of each edition please visit the Idyl E3 home page.

Quick Start

You can quickly get started with Idyl E3 by submitting text through its API. The API documentation contains examples for submitting data using cURL. You can also use one of the available Idyl E3 client SDKs available on our GitHub.


Idyl E3 support multiple entity types and languages. A plugin system allows for consuming text from a variety of sources, extending the extraction capability with additional entity types, and publishing the extracted entities to streams and databases. Extracted entities can be sent to EntityDB for storage and analysis. Idyl E3’s REST API provides endpoints for entity extraction and monitoring health and status.

Input text is processed by specialized files called models. There are sentence models, token models, and entity models. An entity model is a statistical representation of a type of entity for a given language. Examples of possible entity models are:

  • English-language model for persons entities.
  • German-language model for place entities.

You can create your own entity models from your data or you can use any of our pre-created models available on our website. All configuration is done through the file.


There is no cost to install and run Idyl E3. There may be costs associated with plugins, models, and support services and these costs will be outlined on the respective pages.