Visualização de trajetórias acadêmicas: uma abordagem em dados complexos

Abstract

Higher education institutions continuously accumulate extensive volumes of academic data, which constitute valuable resources for understanding the behavior of students, cohorts, and academic programs. However, the growing complexity and scale of these datasets demand innovative approaches to extract meaningful insights and support strategic decision-making, such as optimizing course offerings, reducing dropout rates, and improving pedagogical practices. This dissertation proposes an approach to model students' academic journeys as complex data undergoing temporal evolution, representing them as trajectories in a metric space. Building upon this model, interactive visualization techniques are explored to analyze the academic progression of students across semesters, enabling the identification of recurring patterns, trends, and potential risk situations, such as dropout or delayed graduation. Additionally, the work introduces mechanisms to delimit and refine the data to be visualized through filtering and querying strategies, as well as trajectory summarization techniques that allow the analysis of average behaviors within customizable student groups. The proposed approach was validated through a case study involving real-world data from an undergraduate program, demonstrating its applicability and potential to support comparative analyses and inform educational management processes.