A data-driven approach for the identification of misconceptions in step-based tutoring systems

• Joice Cazanoski Gomes UNISINOS
• Patricia A. Jaques Universidade do Vale do Rio dos Sinos (Unisinos) http://orcid.org/0000-0002-2933-1052

DOI: https://doi.org/10.5753/cbie.sbie.2020.1122

Resumo

Math errors are an important part of the learning process. For this reason, diagnosing them can help teachers and intelligent learning environments to choose the most appropriate type of assistance for the learner. In particular, the identification of learner misconceptions can be of special importance because they represent a misunderstanding of math concepts. In this context, this paper proposes the use of clustering algorithms to automatically identify algebra misconceptions from learners' algebra problem-solving steps in an intelligent learning environment. The computing platform is an intelligent tutoring system that assists students when solving linear equations step by step, by giving minimal and error feedback. The results showed that the model was able to identify some misconceptions already known in the literature, which illustrates the appropriateness of our approach. The automatic identification of misconceptions can help in the identification of new conceptual misunderstanding from large datasets of math problem solving, besides give valuable information for teachers and intelligent learning environments to adapt their instruction and assistance.

Referências

Andersson, J. and Johansson, H. (2015). Using clustering in a cognitive tutor to identify mathematical misconceptions. LU-CS-EX 2015-45.


Elmadani, M., Mathews, M., and Mitrovic, A. (2012). Data-driven misconception discovery in constraint-based intelligent tutoring systems. InProceedings of the 20thInternational Conference on Computers in Education, Singapore.


Feldman, M. Q., Cho, J. Y., Ong, M., Gulwani, S., Popovi ́c, Z., and Andersen, E. (2018).Automatic diagnosis of students’ misconceptions in k-8 mathematics. InProceedings of the 2018 CHI Conference on Human Factors in Computing Systems, page 264, Montreal, Canada. ACM, Author.


Huang, Z. (1998). Extensions to the k-means algorithm for clustering large data sets with categorical values. Data mining and knowledge discovery, 2(3):283–304.


MEC (1997).Parâmetros Curriculares Nacionais 5a a 8a series. MEC.


Russell, M., O’Dwyer, L. M., and Miranda, H. (2009). Diagnosing students’ misconceptions in algebra: Results from an experimental pilot study. Behavior research methods,41(2):414–424.


SAEB (2017).Resultados Avaliação do Ensino Básico. INEP.


VanLehn, K. (2011). The relative effectiveness of human tutoring, intelligent tutoring systems, and other tutoring systems.Educational Psychologist, 46(4):197–221.