PPGEva: sistemas para avaliar a relação entre indicadores e conceitos de PPGs

Abstract

Several graduate courses are offered by universities. In Brazil, these courses are evaluated every four years by CAPES, receiving ratings ranging from three to seven. In order to achieve higher ratings and consequently obtain greater investments, Graduate Programs need to constantly seek improvement. Demonstrating productivity to CAPES is done by improving bibliometric indicators. Since it is not publicly known which indices have the most impact on ratings, this work seeks to systematically identify and classify bibliometric indicators according to their importance. For this purpose, a set of bibliometric data is obtained through computational automation methods and subsequently processed using artificial intelligence techniques until generating a descending classification of the influence of metrics on obtaining higher concepts. A computational system is proposed to facilitate the execution of the process and, thus, assist postgraduate program coordinators in identifying which management plan to use to enhance the rating of their programs. Key-words: Ratings. Bibliometric Indicators. Graduate Programs. Artificial Intelligence.