Authors:
Simone Santos
and
Gilberto S. Junior
Affiliation:
Centro de Informática, Federal University of Pernambuco, Recife, Brazil
Keyword(s):
Computing Education, Student Assessment, Artificial Intelligence, Systematic Literature Review.
Abstract:
This study investigates how Artificial Intelligence (AI) can support student assessment in computing education through a systematic literature review of twenty studies from the past decade. AI’s evolution has significantly impacted various fields, including education, offering advanced capabilities for personalized teaching, continuous evaluation, and performance prediction. Analysing these studies, evidence showed a focus on undergraduate students and the employment primarily face-to-face teaching methods, with engineering education and serious games being more cited contexts. These studies also reveal AI’s potential to create personalized learning experiences using techniques like fuzzy logic, KNN algorithms, and predictive models to analyse student interactions and performance, particularly in educational games and online courses. The positive findings demonstrate AI’s effectiveness in classifying students’ learning profiles, predicting employability, providing real-time assessmen
ts, facilitating targeted interventions, and improving learning outcomes through personalization. Automated assessments via AI have been shown to reduce teachers’ workload by offering accurate, real-time feedback. However, the studies also highlighted challenges concerning student engagement, teacher material quality, model generalization, and technical obstacles such as natural language processing, algorithm stability, and data cleaning. These data-driven factors emphasize the necessity for further advancements in AI to enhance continuous and effective student assessment as part of the personalized learning process.
(More)