Predicting Student Academic Performance using Machine Learning: Analyzing Socio-Economic and Personal Factors from Secondary Education in Portugal

This paper aims to predict student academic performance using machine learning algorithms by analyzing socioeconomic and personal factors from secondary education in Portugal. The study utilizes a dataset containing demographics, family background, educational support, extra-curricular activities, and academic records to evaluate their impact on students' success in Mathematics and Portuguese courses. Machine learning models such as Logistic Regression, Decision Trees, and Random Forests were employed to analyze the data. The Random Forest model outperformed others with an accuracy of 85 %, indicating significant potential in predicting student performance using the selected features.