Semantometria: métrica semântica para calcular o valor de uma publicação por meio de seu texto completo

Abstract

Introduction: The application and use of alternative metrics can be considered as a strategy to be adopted to complement the conventional methods of evaluating scientific production that consider only the citation count. Unlike other metrics that are based on measuring the number of interactions in the scientific communication network, semantometry, the object of study of this research, assumes that the full text is necessary to assess the value of a publication. Given this approach, this article aims to discuss this class of metric, exposing its origin in Herrmannova and Knoth (2014), its analysis methodology and its application formula by means of literature research. Method: this is an exploratory research, in which the preliminary bibliographic survey was conducted adopting seven of the ten initial steps of Query Representation, proposed by Cendón (2018). A narrative literature review was used, with analysis of the recovered documents as a whole. Results: the research on the subject was considered relevant, since there is a great concern with demonstrating the quality of research results to society and it was found that there are few publications on the subject, materialized by the results of searches in databases: Dimensions, Gale, Scopus and Web of Science. Conclusions: it is concluded that, although semantometry has advantages in its application, it is necessary to continue studying this field to better understand which facets of research quality they can capture and how they can best be applied. It was also observed its association with the mining and analysis of textual data and its relationship with techniques and tools of textual similarities, as well as with research involving altmetrics.