Why RosaeNLG, the author etc.

This is the documentation for 1.16.6 version, which is not the latest version. Consider upgrading to 3.4.0.

Commercial systems

NLG has existed for a long time as an academic subject but it is only recently that commercial NLG technology has become widely available and self service.

World-class NLG actors are:

  • Narrative Science

  • Yseop

  • Arria NLG

  • Automated Insights

  • AX Semantics

(but there are some other actors here too - if you represent one, contact me and I’ll add you to the list)

Each Natural Language Generator has its own specificities, and you should compare before choosing one.

Open-source NLG?

There is some open source for NLG but it is generally:

  • Completely outdated or not maintained.

  • It focuses on one very specific NLG feature (and does it well) but is not complete enough to build real life projects, e.g. SimpleNLG that only adresses surface realisation.

So as you might have guessed I decided to write my own Natural Language Generator, and make it open-source.

RosaeNLG’s characteristics as a natural language generator

There are various techniques to generate texts. Template based generators use templates, which are a mix of static content (plain text) and dynamic content. Think of PHP etc.

In template-based system, most of the time you don’t really care about the exact structure of the text (subject, verb, etc.). You don’t need to be a linguist to use them: you only need a quite basic understanding of the output language grammar.

The characteristics of RosaeNLG are:

  • template-based

  • easy to use

  • based on modern & mature technologies: JavaScript & pug

  • complete enough to build real life projects

  • open-source of course

Why is it named RosaeNLG

Rosa means rose in Latin, and rosae is the plural of rosa. It is a reference to the linguistic complexities hidden in each language.

The author

You may contact me at ludan [dot] stoeckle [at] rosaenlg [dot] org