Modelling Students in Mathematical; A Data Mining Approach for Early Intervention 

Teacher mastery and student willingness have always been seen as pivotal to academic success in mathematics. However, emerging research highlights an even deeper influence - students' mathematical mindset - as critical foundation for unlocking their full potential. This study aimed to construct a predictive model to identify students' mathematical mindsets using Educational Data Mining (EDM). Three classification algorithms – Naive Bayes (NB), Support Vector Machine (SVM), and Multilayer Perceptron Neural Network (MLP) – were implemented on a dataset of 633 junior high school students. The dataset that was collected induded the students' demographic information, academic performance, and responses to validated questionnaires on mathematics anxiety, time management, and mindset. Among the three algorithms, SVM demonstrated superior performance with an accuracy of 98.8%, significantly outperforming Naive Bayes (95.2%) and Multilayer Perceptron Neural Network (93%). These findings highlighted the potential of EDM, particularly SVM, to accurately predict students' mindsets, enabling educators to implement targeted interventions and foster a growth mindset culture.

Keywords: Educational Data Mining, Naive Bayes, Support Vector Machine, Multilayer Perceptron Neural Network, Students' Mindset in Mathematics

Available at: SRCB College Library