Diretrizes para a indexação semiautomática de publicações científicas por meio de sintagmas nominais

Abstract

This study involves aspects concerning semi-automatic indexing and nominal syntagms intending to contribute to information representation and retrieval practices and research in the field of Information Science and related areas. From this point of view, this research aims to propose guidelines for the semi-automatic indexing of scientific productions through nominal syntagms., considering the theoretical-methodological background of manual indexing and semi-automatic indexing and the potential of Nominal Syntagms. For that, semiautomatic indexing strategies were identified in the national and international scientific literature; the composition and potential of nominal syntagms as information-carrying units were investigated; the functioning of the PALAVRAS software in the identification/extraction of nominal syntagms was analyzed through the analysis of scientific articles in the area of Thematic Representation of Information; and procedures were outlined for the semi-automatic identification of NSs that function as descriptor elements of the domain of Thematic Representation of Information. The methodology used was bibliographical, documentary, exploratory and descriptive research with a qualitative and quantitative approach. The research comprises the scientific articles presented in TG2 of the XXII National Meeting of Research and Postgraduate Studies in Information Science, 2022 edition. The theoretical and methodological support was sought in the semantic potential of nominal syntagms applied in semi-automatic indexing as units that carry precise meanings and are viable in the thematic representation of information. As a result, it was highlighted that the semi-automation of indexing associated with the use of nominal syntagms offers a feasible alternative in the context of increasing information production in a digital environment and the semantic richness of nominal syntagms was evidenced, which can be used as access points to the document in a more agile and objective way, providing a reducing indexing time, with the help of automatic information processing, without freeing the human eye in a task that is subjective by nature. It was also found that the association of semi-automatic indexing with the use of nominal syntagms proved to be a promising alternative for document indexing since it proposes guidelines that seek to guide the construction of semi-automatic indexing methodologies that use the potential of nominal syntagms and presents a path for the semiautomatic thematic analysis of texts, based on pre-established and validated criteria, to achieve, in a short space of time, representative document descriptors of documents in information units and information retrieval systems. Hence, the association of human indexing with automation and the usage of nominal syntagms with information-bearing units proves to be a promising and effective alternative for the thematic representation of information in the digital age.