Classificação de Conjuntos de Dados de Investigação com Base em seus Registos de Metadados

Abstract

As important as accumulating knowledge, it is essential to ensure that it will be available when needed, in an effective and efficient manner. The retrieval of information in the digital environment presents innovative opportunities and challenges for information democratization. Open access, more than a philosophy, has become a public demand. This dissertation studies the use of classifications as a way of retrieving digital information in data repositories. The work proposes the construction of a classifier of documents from textual metadata such as title, keywords and description of digital objects. Classification systems and thesauri are used, classic structures of Information Science as support or subsidy to the developed model. In order to validate the classifier a crowdsourcing tool is presented,with a strong emphasis on the interoperability of the solution. A portal was constructed in order to demonstrate, in a practical way, the results of the experiments that were carried out. It also approaches the paradigm of information retrieval from the classification of digital objects with the goal of information democratization and literacy. The tool is applicable to both organizations and individuals.