What is NLUlite?

NLUlite is a developer-friendly database that can read texts and answer questions about them, by combining a natural language parser and a graph database. For example, consider the following text (taken from http://simple.wikipedia.org/wiki/Snakes)

A snake is a member of about 19 reptile families that has no limbs,
voice, ears, or eyelids. They also have a long, slender body. About
2,900 snake species are known to exist, and most of them live in the
tropics. Very few snake species are able to live beyond the Tropic of
Cancer or Tropic of Capricorn, and only one species, common viper,
lives beyond the Arctic Circle. Their skin is covered with
scales. They can see well, and they can taste the air with their
tongues by flicking them in and out. Though they do not have a voice,
they hiss instead. Like all reptiles, snakes need the heat of the sun
to control their body temperature.

You can retrieve the information in the text by asking questions that match the sentence within the provided data

what are the snakes able to do?
where do most of the snakes live?
what animal has no limbs?
what is a member of what?

You can also combine two or more questions

what has no voice and sees well?

Additionally, you can specify in the text simple inference rules, like

If an animal has no limbs it cannot walk.

and then ask:

what does not walk?

The answer will be computed in terms of the text you submitted and the rules you provided. All the code examples in this tutorial are provided in the path NLUlite/example_python/ in the installation.

You are welcome to try out the online example. A more technical descriptions of the inner workings of NLUlite can be found in these slides.