Detection of learning styles with prior knowledge data using the SVM, K-NN and Naïve Bayes algorithms

Hasibuan, M. Said and Aziz, RZ Abdul (2022) Detection of learning styles with prior knowledge data using the SVM, K-NN and Naïve Bayes algorithms. JURNAL INFOTEL, 14 (3). pp. 209-213. ISSN 2085-3688

Text Detection of learning styles with prior knowledge data.pdf Download (935kB)

Abstract

Data-driven (DD) and literature-based techniques for autonomous learning style detection are the two categories (LB). Both techniques of automatic learning style detection offer advantages over traditional learning style detection methods because they leverage external data sources that are more accurate than surveys in traditional styles of detection, such as forums, quizzes, and views of teaching materials. On the other hand, the results of automatic detection do not always reflect learning styles. To tackle these issues, this work provides a learning style recognition algorithm that draws on data from the learner's internal source, namely past knowledge. Prior knowledge is advocated because it is based on the learner's knowledge or skills, which better reflect the learner's traits, rather than the learner's dynamic behaviour. This research proposes a method for recognising autonomous learning patterns that rely on prior information. The learning style detection framework is unusual in that it has three stages: prior knowledge question formulation, prior knowledge measurements, and learning style detection utilising SVM, Naïve Bayes, and K-Nearest Neighbour (K-NN) classification algorithms. The results showed that Naïve Bayes has an accuracy value of 91.48%, K-NN of 89.39% and SVM of 87.31%.

Item Type:	Article
Subjects:	Ilmu Komputer 600 Teknologi - Ilmu terapan > 620 Ilmu Teknik - Engineering
Divisions:	Artikel Ilmiah Dosen
Depositing User:	Dr. RZ Abdul Aziz
Date Deposited:	07 Oct 2022 04:06
Last Modified:	07 Oct 2022 04:06
URI:	http://repo.darmajaya.ac.id/id/eprint/9729

Actions (login required)

View Item