Converting ontologies into DSLs

• Main Author: Fonseca, João Manuel de Sousa
• Publication Date: 2014
• Format: Master thesis
• Language: eng
• Source: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Download full: http://hdl.handle.net/1822/36786
• topic: Ontologies; DSL; Domain-specific languages; 681.3.06; Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

This paper presents a project whose main objective is to explore the Ontological-based development of Domain Specific Languages (DSL), more precisely, of their underlying Grammar. After reviewing the basic concepts characterizing Ontologies and Domain-Specific Languages, we introduce a tool, OWL2Gra, that takes profit of the knowledge described by the ontology and automatically generates a grammar for a DSL that allows to discourse about the domain described by that ontology. This approach represents a rigorous method to create, in a secure and effective way, a grammar for a new specialized language restricted to a concrete domain. The usual process of creating a grammar from the scratch is, as every creative action, difficult, slow and error prone; so this proposal is, from a Grammar Engineering point of view, of uttermost importance. After the grammar generation phase, the Grammar Engineer can manipulate it to add syntactic sugar to improve the final language quality or even to add semantic actions. The OWL2Gra project is composed of three engines. The main one is OWL2DSL, the component that converts an OWL ontology into an attribute grammar. The two additional modules are Onto2OWL and Ddesc2OWL. The former, Onto2OWL, converts ontologies written in OntoDL (a light-weight DSL to describe ontologies) into standard OWL XML that can be loaded into the well known Proteg´ e sys- ´ tem to future editing; the later, Ddesc2OWL, converts domain instances written in the DSL generated by OWL2DSL into the initial OWL ontology. Ddesc2OWL plays an important role because it allows for the population of the original ontology with concept and relation instances extracted from the new language concrete sentences this allow a faster ontology population.