AI in evaluating software courses: a pilot undergraduate thesis tests support for teachers
Caio César Oliveira defended at FCTE/UnB an undergraduate thesis developing artificial intelligence to improve individual evaluations in Software Engineering courses.
Brasília, September 12, 2024
Individual evaluation in Software Engineering courses is a double task: care for pedagogical fairness and respond to scale. The undergraduate thesis defended in September 2024 by Caio César Oliveira proposes AI as a support to the instructor in that job, under the advising of Prof. George Marsicano Corrêa .
The work departs from a real scenario: the Software Requirements and Software Development Methods courses at FCTE/UnB, whose assessments run through Microsoft Forms. Around that funnel, the author designs an intelligent assistant that helps the instructor organize responses, identify patterns, flag labels worth reviewing, and — when applicable — propose grading priorities.
By tying AI to concrete pedagogical practice , the thesis enters a discussion the academic community cannot postpone: the role of intelligent assistants in evaluation, without letting judgment slip inside a black box. It is a pilot work — and useful precisely for that — sketching a solution that can be tested, adjusted, and critiqued in use.
The full text — written in Portuguese — is available at the UnB Undergraduate Theses Digital Library.
About CEDIS: The Center for Studies, Development, and Innovation in Software (CEDIS), linked to the University of Brasília, researches and develops innovative software solutions.