2022/09/12 by Ahmad Mousa Altamimi, Altamimi, Ahmad Mousa, Mohammad Azzeh +3
Computer Science · Psychology · #Computers and Society (cs.CY) #FOS: Computer and information sciences #Learning Styles and Cognitive Differences #Machine Learning (cs.LG) #Online Learning and Analytics
paper · pdf · doi:10.48550/arxiv.2209.12691
openalex publication_date 2022/09/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Traditional learning systems have responded quickly to the COVID pandemic and moved to online or distance learning. Online learning requires a personalization method because the interaction between learners and instructors is minimal, and learners have a specific learning method that works best for them. One of the personalization methods is detecting the learners' learning style. To detect learning styles, several works have been proposed using classification techniques. However, the current detection models become ineffective when learners have no dominant style or a mix of learning styles. Thus, the objective of this study is twofold. Firstly, constructing a prediction model based on regression analysis provides a probabilistic approach for inferring the preferred learning style. Secondly, comparing regression models and classification models for detecting learning style. To ground our conceptual model, a set of machine learning algorithms have been implemented based on a dataset collected from a sample of 72 students using visual, auditory, reading/writing, and kinesthetic (VARK's) inventory questionnaire. Results show that regression techniques are more accurate and representative for real-world scenarios than classification algorithms, where students might have multiple learning styles but with different probabilities. We believe that this research will help educational institutes to engage learning styles in the teaching process.