Recomendações para uma ontologia de domínio no âmbito de objetos digitais a luz da Ciência da Informação

Abstract

Different processes that make up the Knowledge Organization undergo continuous transformations. They are the ones that tend to accompany everything from communication models to technologies associated with the production, registration, transmission, access and use of knowledge. In Information Science, Knowledge Organization appears as a sub-area of research, but in the scientific environment, the domains of the latter go beyond the former and are concentrated in a consolidated manner in the areas of Computing and, mainly, in Knowledge Engineering, above all, in applied mode, through Ontologies. The present thesis takes as a premise the context in question and the belief that, through the basic principles of Ontologies in Information Science and machine learning in Artificial Intelligence, it is possible to establish recommendations for a Domain Ontology in the context of Imagetic Digital Objects under the perspective of social relevance in systems that act intelligently, enabling other organizational strategies for accessing and retrieving data and documents. As a general objective, recommendations were elaborated for a Domain Ontology within the scope of Imagetic Digital Objects in the light of Information Science. As specific objectives, bibliographies on Ontologies and Knowledge Organization Systems in Information Science were analyzed; mapped bibliographies on modeling domains using images in Information Science; raised principles for the development of Ontologies in Information and Computing Science; in addition to presenting parameters for image processing that enable ontological modeling. The investigative method is characterized as mediate deductive and the research, from the point of view of the objectives, is typified as explanatory, with a quantitative and qualitative approach, carrying out bibliographic procedures and content analysis. As a result, 10 (ten) recommendations are presented to generate a Domain Ontology using the context of digital images under the aegis of the Organization of Knowledge in Artificial Intelligence environments, intelligible to humans, and processable by machine. It is concluded that the modeling of domains, based on the use of imagery language, promotes innovation, accessibility and increment to representational means and recovery of knowledge records.