Um modelo para detecção automática do comportamento de tentativa e erro em STI baseado em passo
Resumo
Este artigo apresenta um modelo baseado em aprendizado de máquina para detecção automática do comportamento de tentativa e erro em um Sistema Tutor Inteligente baseado em passos. Esse modelo foi treinado com base nos dados obtidos através da observação e análise de vídeos e dos logs de interação do aluno com o sistema. Após o desenvolvimento de um protocolo de anotação, rotulamos as ações dos alunos com a ocorrência ou não do comportamento de abuso de tentativa e erro. Submetemos o modelo a dados não usados previamente, nos quais o modelo alcançou um Kappa de 0,684.
Palavras-chave:
Sistemas Tutores Inteligentes, Comportamento de Tentativa e Erro, Mineração de Dados Educacionais, Gaming The System
Referências
Aleven, V., McLaren, B., Roll, I., and Koedinger, K. (2004). Toward tutoring help seeking. In Intelligent Tutoring Systems, pages 227–239, Berlin. Springer.
Azevedo, O., de Morais, F., and Jaques, P. A. (2018). Exploring gamification to prevent gaming the system and help refusal in tutoring systems. In EC-TEL, pages 231–244.
Baker, R. and et al. (2013). Modeling and studying gaming the system with educational data mining. In Int. Handbook of Metacog. and Learn. Techn., pages 97–115. Springer.
Baker, R. S. (2006). Designing Intelligent Tutors That Adapt to when Students Game the System. PhD thesis, Carnegie Mellon University, Pittsburgh, PA, USA. AAI3241593.
Chawla, N., Bowyer, K., Hall, L., and Kegelmeyer, W. (2002). Smote: Synthetic minority oversampling technique. J. Artif. Intell. Res. (JAIR), 16:321–357.
Jaques, P. A. and et al. (2013). Rule-based expert systems to support step-by-step guidance in algebraic problem solving: The case of the tutor pat2math. Expert Systems With Applications, 40(14):5456–5465.
Landis, J. and Koch, G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1):159–174.
Morais, F. and et al. (2019). Em AP-ML: A Protocol of Emotions and Behaviors Annotation for Machine Learning Labels. In EC-TEL, Netherlands. Springer.
Ocumpaugh, J. (2015). Baker rodrigo ocumpaugh monitoring protocol (bromp) 2.0 technical and training manual.
Paquette, L. and et al. (2018). A system-general model for the detection of gaming the system behavior in ctat and learnsphere. In AIED, pages 257–260. Springer.
Peters, C. and et al. (2018). Predictors and outcomes of gaming in an intelligent tutoring system. In ITS, pages 366–372. Springer.
Randolph, J. J. (2005). Free-marginal multirater kappa (multirater k [free]): An alternative to fleiss’ fixed-marginal multirater kappa.Online submission.
Santos, M., Soares, J., Abreu, P., Araujo, H., and Santos, J. (2018). Cross-validation for imbalanced datasets: Avoiding overoptimistic and overfitting approaches.IEEE.
Seffrin, H. and et al. (2012). Dicas inteligentes no sistema tutor inteligente pat2math. In Simpósio Brasileiro de Informática na Educação (SBIE), volume 23.
VanLehn, K. (2011). The relative effectiveness of human tutoring, intelligent tutoring systems, and other tutoring systems. Educational Psychologist, 46(4):197–221.
Woolf, B. P. (2007). Building Intelligent Interactive Tutors: Student-centered Strategies for Revolutionizing e-Learning. Morgan Kaufmann.
Azevedo, O., de Morais, F., and Jaques, P. A. (2018). Exploring gamification to prevent gaming the system and help refusal in tutoring systems. In EC-TEL, pages 231–244.
Baker, R. and et al. (2013). Modeling and studying gaming the system with educational data mining. In Int. Handbook of Metacog. and Learn. Techn., pages 97–115. Springer.
Baker, R. S. (2006). Designing Intelligent Tutors That Adapt to when Students Game the System. PhD thesis, Carnegie Mellon University, Pittsburgh, PA, USA. AAI3241593.
Chawla, N., Bowyer, K., Hall, L., and Kegelmeyer, W. (2002). Smote: Synthetic minority oversampling technique. J. Artif. Intell. Res. (JAIR), 16:321–357.
Jaques, P. A. and et al. (2013). Rule-based expert systems to support step-by-step guidance in algebraic problem solving: The case of the tutor pat2math. Expert Systems With Applications, 40(14):5456–5465.
Landis, J. and Koch, G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1):159–174.
Morais, F. and et al. (2019). Em AP-ML: A Protocol of Emotions and Behaviors Annotation for Machine Learning Labels. In EC-TEL, Netherlands. Springer.
Ocumpaugh, J. (2015). Baker rodrigo ocumpaugh monitoring protocol (bromp) 2.0 technical and training manual.
Paquette, L. and et al. (2018). A system-general model for the detection of gaming the system behavior in ctat and learnsphere. In AIED, pages 257–260. Springer.
Peters, C. and et al. (2018). Predictors and outcomes of gaming in an intelligent tutoring system. In ITS, pages 366–372. Springer.
Randolph, J. J. (2005). Free-marginal multirater kappa (multirater k [free]): An alternative to fleiss’ fixed-marginal multirater kappa.Online submission.
Santos, M., Soares, J., Abreu, P., Araujo, H., and Santos, J. (2018). Cross-validation for imbalanced datasets: Avoiding overoptimistic and overfitting approaches.IEEE.
Seffrin, H. and et al. (2012). Dicas inteligentes no sistema tutor inteligente pat2math. In Simpósio Brasileiro de Informática na Educação (SBIE), volume 23.
VanLehn, K. (2011). The relative effectiveness of human tutoring, intelligent tutoring systems, and other tutoring systems. Educational Psychologist, 46(4):197–221.
Woolf, B. P. (2007). Building Intelligent Interactive Tutors: Student-centered Strategies for Revolutionizing e-Learning. Morgan Kaufmann.
Publicado
24/11/2020
Como Citar
CASTILHOS, Fábio M.; MORAIS, Felipe; AZEVEDO, Otávio Bastos; JAQUES, Patricia A..
Um modelo para detecção automática do comportamento de tentativa e erro em STI baseado em passo. In: SIMPÓSIO BRASILEIRO DE INFORMÁTICA NA EDUCAÇÃO (SBIE), 31. , 2020, Online.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2020
.
p. 1062-1071.
DOI: https://doi.org/10.5753/cbie.sbie.2020.1062.
