A profile analysis of the top Brazilian Computer Science graduate programs

Abstract

Analyzing the research productivity of a country, an academic institution or even a single research group contributes to understand how science evolves and discovers new research perspectives, since such efforts usually reveal key aspects that can be improved, avoided or even applied to other contexts. In this article, we present a detailed analysis of the top Brazilian Computer Science graduate programs. The analysis involves profile data on faculty members (e.g., career length and number of mentored students) and on the quality of their research efforts, assessed using the quality of their publications and collaboration patterns. The objective is to uncover factors that explain the strengths and weaknesses of graduate programs. Results show that the highest ranked programs include more experienced faculty members, who have mentored more Ph.D. students. We also show that programs target distinct publication venues, with the best ranked ones focusing on higher quality conferences and journals. By analyzing collaboration patterns, we show that intra-program relationships occur quite naturally whereas inter-program ones are still very incipient.