Machine Learning Bias in Predicting High School Grades: A Knowledge Perspective

This study focuses on the machine learning bias when predicting teacher grades. The experimental phase consists of student grades 11th and 12thgrade Portuguese high school computing variance decomposition. In base implementation, only academic achievement critical factors are considered. second preceding year’s grade is appended as an input variable. algorithms in use random forest, support vector machine, extreme boosting machine. reasons behind poor performance either space preciseness or lack a sound record performance. We introduce new concept knowledge predictive model classification. Precision education would reduce by providing low-bias intensive-knowledge models. To avoid bias, it not necessary to add space. Low-bias extensive-knowledge models achievable simply appending student’s earlier model. promoted precision suited designing policies actions toward attainments. If aim solely prediction, deciding for low knowledge-extensive can be appropriate correct. Doi: 10.28991/esj-2021-01298 Full Text: PDF