Project scope

A formal representation of Esperanto’s grammar

This project comes from the question: which could be the best natural language for vocal interaction between humans and computers?

Esperanto was therefore considered, an artificial language officially born in 1887 for pleasure and thanks to the work of Polish Doctor Ludwik Lejzer Zamenhof, a passionate linguistics studier that wanted to develop a language purely international. From this, Esperanto was adopted by a significant community of speakers in all the world really able to maintain a serious conversation in this language. The choice of Esperanto was thanks simplicity and regularity of its grammar that could makes easier for computing systems to manage the structure of the language.

The project “Ontology of Esperanto” is proposed as one of the first attempts to produce a formal representation of a non formal language (natural language), based on the Semantic Web standards.

The goal was to develop an ontology usable into NLP software tools that can make easy a computer understand the logical form of varying complexity phrases expressed in Esperanto. From here we can imagine a scenario which on the Semantic Web becomes available innovative services that allow users perform queries in natural language for accessing data and documents, without having any expertise in formal query languages.

The OWL-DL ontology of Esperanto

Ontology is written in OWL-DL, a version of the language recommended by the W3C for the Semantic Web, using Protégé, the ontology editor developed by the University of Stanford (US), that uses the automated reasoning services of Racer, the semantic reasoner developed by the University of Luebeck (D).

The ontology verily are three: first for the grammar analysis, second for the logical analysis and thitd for the analysis of the sentence, where the highest level ontology refers to concepts at lower levels. In this way, once the various parts of the text is qualified at lower levels, so the higher levels can be automatically inferred.