Modelagem de domínios em Sistemas de Organização do Conhecimento (SOC): uma investigação em tesauros e ontologias para a informação legislativa

Abstract

The Knowledge Organization Systems (KOS) are instruments that deal with document content representation for storage, organization, retrieval, and sharing purposes. The elaboration of SOCs implies in modeling, in other words, a creation of semantic models that are simplified descriptions of the reality. One relevant domain is the legal domain,which, in addition to study the production processes of legal norms for human coexistence in society, also produces legislative information. The present research is justified by the need of gathering studies about modeling from the two current theoretical-methodological slants in Information Science, in order to contribute to the creation of a SOC aligned to the needs of legislative information. The main objective isto investigate domains modeling from the theoretical-methodological framework of both thesauri and ontologies for legislative information. Specific objectives include: i) gathering theoretical and methodological foundations on domain modeling; ii) to characterize brazilian legal domain and the specificities of legislative information; iii) to develop a prototype of thesaurus for legislative information, particularly, on MedicalRights; iv) to develop a prototype of an ontology for a legislative information on the same domain; v) to compare and to analyze domain modeling from the construction of thesauri and of ontologies. The research is classified as qualitative, employing the case study as the technical procedure (construction of prototypes). As results, we present acomparative study of domain modeling processes in thesauri and ontologies, as well as a synthesis of our observations. The synthesis of the observations consisted in diagrams and synoptic tables with the operations and general principles of the domain modeling in thesauri and ontologies, a generalizable model and applicable not only tothe legal documentation. The study allowed us to conclude that both methodologies approach similarly the domain modeling activity in creating semantic models based on concepts, definitions and semantic relations. It was also verified that the theoreticalmethodologicalreferences of thesauri and ontologies are complementary with respectto domain modeling.