Predicting Students' Performance in Zimbabwean Secondary Schools Using Machine Learning Techniques: A Case of Glenview Mufakose District

ABSTRACT This study investigates the potential of machine learning techniques to predict student academic performance in Zimbabwean secondary schools, focusing specifically on the Glen View and Mufakose district. Recognizing the critical need for early identification of at-risk students to facilitate timely interventions and improve educational outcomes, there has been need to develop a predictive model for this purpose. Several machine learning techniques such as Random Forest, XGBoost, Support Vector Machines (SVM), Decision Trees, and Neural Networks, logistic regression will be examined because they have been showing promising results as they were used in other educational contexts. Each model will be evaluated using standard performance criteria like accuracy, precision, recall, and F1-score, and consideration made in favour of that with the greatest accuracy. The idea is to find a model that best fits Zimbabwe's unique secondary school environment. Once the model has been developed, educators, parents and policymakers can make valuable use of the predictions to help assist at risk student and develop targeted intervention. Key words: MOPSE, Educator, Learner, MLA, Policy makers, Institutions of learning, Correlation, Algorithm, SVM, MLP, XGBoost, SMOTE.