Integrando Dados Textuais e Estruturados para Compreender Fatores Associados à Infrequência Escolar
Resumo
School dropout is influenced by both academic and socio-demographic factors, yet Educational Data Mining (EDM) studies predominantly rely on structured quantitative data. As a result, contextual information contained in textual records remains underexplored. This paper proposes a Knowledge Discovery in Databases (KDD) process that integrates unstructured and structured data from FICAI, a school absenteeism monitoring system used in southern Brazil. The approach applies topic modeling to reports produced by child protection counselors to identify latent topics associated with school absenteeism. These topics are then combined with demographic and institutional attributes and analyzed through Association rules. The study analyzes records from approximately 89,800 students. The results show that text mining uncovers factors overlooked by traditional institutional diagnoses. The discovered associations indicate that the pedagogical triad reflects symptoms rather than causes of disengagement, reveal distinct risk profiles across educational stages, and provide evidence to support targeted dropout prevention policies.
Referências
Chen, X., Zou, D., Cheng, G., and Xie, H. Detecting Latent Topics and Trends in Educational Technologies Over Four Decades Using Structural Topic Modeling: A Retrospective of all Volumes of Computers & Education. Computers & Education vol. 151, pp. 103855, 2020.
de Jesus, J. A. and de Gusmão, R. P. Investigação da Evasão Estudantil por meio da Mineração de Dados e Aprendizagem de Máquina: Um Mapeamento Sistemático. Revista Brasileira de Informática na Educação vol. 32, pp. 807–841, mar., 2024.
Dol, S. M. and Jawandhiya, P. M. Systematic Review and Analysis of EDM for Predicting the Academic Performance of Students. Journal of The Institution of Engineers (India): Series B 105 (4): 1021–1071, 2024.
Han, J., Kamber, M., and Pei, J. Data Mining: Concepts and Techniques. Morgan Kaufmann, Waltham, MA, 2011.
Haque, U. M., Kabir, E., and Khanam, R. Investigating school absenteeism and refusal among Australian children and adolescents using Apriori association rule mining. Scientific Reports vol. 14, pp. 1443, 2024.
Hossain, M. and Mondal, M. N. I. Mining Hidden Patterns in Student Dropout: Using Random Forest-Based Feature Selection and Association Rule Mining. In Proceedings of the 2nd International Conference on Next-Generation Computing, IoT and Machine Learning (NCIM). IEEE, New York, NY, USA, pp. 1–6, 2025.
Kastrati, Z., Imran, A. S., and Kurti, A. Weakly Supervised Framework for Aspect-Based Sentiment Analysis on Students’ Reviews of MOOCs. IEEE Access vol. 8, pp. 106598–106610, 2020.
Silva Filho, R. B. and de Lima Araújo, R. M. Evasão e abandono escolar na educação básica no Brasil: fatores, causas e possíveis consequências. Educação por Escrito 8 (1): 35–48, 2017.
