Exploring students' performance in mathematics in Portugal using data analytics techniques: a data science use-case

• Main Authors: Hassan, Raini, Fadzleey, Nur Zulfah Insyirah, Ab Hamid, Annesa Maisarah, Abd Aziz, Rabiatul Adawiyah, Jamalullain, Afiefah, Syaiful 'Adli, Fatin Syafiqah
• Format: Book Chapter
• Language: English
• Published: KICT Publishing 2024
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://irep.iium.edu.my/112235/1/112235_Exploring%20students%27%20performance%20in%20mathematics.pdf; http://irep.iium.edu.my/112235/; https://kulliyyah.iium.edu.my/kict/fyp-ebook-adict/
• Institution: Universiti Islam Antarabangsa Malaysia

This research project investigates the relationship between family background and student performance in Mathematics in Portugal. The analysis is based on an open- source dataset from Kaggle Datasets, comprising 395 rows and 33 columns, with 24 key features used for predictive analysis. The purpose is to identify the key factors influencing academic performance, providing insights for targeted interventions and support systems. Machine learning algorithms, specifically Random Forest Regression and Decision Trees, are utilized to analyze the dataset and determine the most significant factor impacting student performance. The study employs descriptive and predictive analytics techniques to understand student performance patterns and forecast future outcomes based on family background factors. The practical application of this research lies in developing predictive models that inform data- driven decisions by educators and policymakers. The results, as shown in Table III, indicate that the Random Forest Regression model outperforms the Decision Tree model, achieving lower Mean Squared Error (9.6212), Root Mean Squared Error (3.0842), and Mean Absolute Error (2.4060). The findings highlight the importance of parental education levels and positive family relationships in influencing academic performance in Mathematics. Future research endeavours should explore the applicability of these findings to other nations, such as Malaysia, to gain a broader understanding of the factors influencing student academic success and adapt data-driven interventions accordingly.