KNearest Oracle-AutoML Model for Predicting Student Dropouts in Tanzanian’s Secondary Schools
Keywords: Machine Learning Techniques, Automated Machine Learning, Hyper-Parameter Optimization Techniques, Student Dropout Prediction
Abstract: Secondary school dropout is a major problem in developing countries, particularly in Sub-Saharan Africa. In Tanzania, student dropouts in secondary schools increased from 3.8 percent in 2018 to 4.2 percent in 2019. Student dropout rates increased significantly in secondary schools due to inappropriate identification of the root causes of student dropouts and the method used to project the severity of the problem. In addressing this prevalent problem, machine learning is designed to learn from data, revealing previously unknown findings as it discovers historical relationships and trends. The proposed model has done well in addressing secondary school dropouts by accurately identifying the root causes of student dropout. This study discovered that the root causes of student dropout in Tanzanian secondary schools are the number of children, household size, distance, age, household education, student location (area), student gender, and means to the
school. Therefore, the enhanced prediction scores indicate an accurate selection of student dropout features that significantly contribute to student dropout, which can be closely examined during the learning process to allow for early interventions.
Submission Category: Machine learning algorithms
Submission Number: 6
Loading