Clustering for identifying question levels
Clustering is an unsupervised data partitioning method that aims to divide the dataset according to characteristics intrinsic to each element, satisfying some criteria such that elements of the same cluster are more similar than those in different ones [1]. Among the unsupervised methods, clustering techniques are the most popular. Due to its versatility, the clustering procedure is very useful in engineering, health sciences, humanities, economics, education, and other areas.
Example:
The work [2] investigates, through the application of two clustering methodologies, the optimal number of difficulty levels to reorganize the questions of the MathE platform.
Reference:
[1] Aggarwal, C. & Reddy, C. Data Custering Algorithms and Applications. Taylor & Francis Group. (CRC Press,2013 )
[2] Azevedo, B.F., Amoura, Y., Rocha, A.M.A.C., Fernandes, F.P., Pacheco, M.F., Pereira, A.I. (2022 ). Analyzing the MathE Platform Through Clustering Algorithms. In: Gervasi, O., Murgante, B., Misra, S., Rocha, A.M.A.C., Garau, C. (eds ) Computational Science and Its Applications – ICCSA 2022 Workshops. ICCSA 2022. Lecture Notes in Computer Science, vol 13378. Springer, Cham. https://doi.org/10.1007/978-3-031-10562-3_15
[2] Azevedo, B.F., Amoura, Y., Rocha, A.M.A.C., Fernandes, F.P., Pacheco, M.F., Pereira, A.I. (2022 ). Analyzing the MathE Platform Through Clustering Algorithms. In: Gervasi, O., Murgante, B., Misra, S., Rocha, A.M.A.C., Garau, C. (eds ) Computational Science and Its Applications – ICCSA 2022 Workshops. ICCSA 2022. Lecture Notes in Computer Science, vol 13378. Springer, Cham. https://doi.org/10.1007/978-3-031-10562-3_15
Author of the tip:
Florbela P. Fernandes
Instituto Politécnico de Bragança