Mathvision prototype using predictive analytics / Yuzi Mahmud ... [et al.]

Mahmud, Yuzi and Abdul Razak, Muhd Syahir and Abdul-Rahman, Shuzlina and Hanafiah, Mastura and Suhaimi, Amien Ashraf (2023) Mathvision prototype using predictive analytics / Yuzi Mahmud ... [et al.]. Malaysian Journal of Computing (MJoC), 8 (2): 4. pp. 1505-1516. ISSN 2600-8238


Malaysia is currently going towards Industrial Revolution (IR) 4.0 which makes Science, Technology, Engineering and Mathematics (STEM) subjects become more crucial. IR 4.0 covers a lot of aspects especially in digital transformation in manufacturing, and this certainly requires strong mathematical knowledge. To achieve this goal, students need to have a good foundation in Mathematics subject. However, due to the increased number of students nowadays, teachers are facing challenges to track students’ progress efficiently. In this study, a predictive model has been developed that aims to assist Mathematics teachers in monitoring their students. The prototype, called MathVision, can track students’ progress effectively in each topic and subtopic of Mathematics subject and predict the grades that students will obtain based on the history result. A total of 207 instances was collected among Form 5 students from a government school to represent the samples for the modelling task. The Multiclass Decision Forest algorithm appeared to be the best predictive model with 95.16% accuracy, as compared to Boosted Decision Tree, Logistic Regression, and Neural Network. Flutter framework and Firebase services were used for front-end and back-end system respectively, and Microsoft Power BI was used for data visualization. The result of prototype testing showed that MathVision could predict students’ grade for Quiz 2 based on Quiz 1 performance. MathVision is also capable for real-time prediction that guarantees an immediate response time which can help Mathematics teachers to support students who need further assistance in this subject based on the prediction given. For MathVision’s future improvement, the number of instances needs to increase, and more significant variables need to be added.


Item Type: Article
Email / ID Num.
Mahmud, Yuzi
Abdul Razak, Muhd Syahir
Abdul-Rahman, Shuzlina
Hanafiah, Mastura
Suhaimi, Amien Ashraf
Subjects: Q Science > QA Mathematics > Mathematical statistics. Probabilities > Prediction analysis
Divisions: Universiti Teknologi MARA, Shah Alam > Faculty of Computer and Mathematical Sciences
Journal or Publication Title: Malaysian Journal of Computing (MJoC)
UiTM Journal Collections: UiTM Journal > Malaysian Journal of Computing (MJoC)
ISSN: 2600-8238
Volume: 8
Number: 2
Page Range: pp. 1505-1516
Keywords: Data analytics, IR 4.0, mathematics, predictive model, real-time prediction
Date: October 2023
Edit Item
Edit Item


[thumbnail of 86382.pdf] Text

Download (556kB)

ID Number




Statistic details