Establishment of pass/fail prediction model based on Support Vector Machine

Abstract

If the transition of academic ability of individual students can be measured by a specific index, it will lead to the provision of better education for students and an accurate evaluation of the effectiveness of provided education. Therefore, the main purpose of this paper is to examine whether pass/fail prediction of an examination derived from a Support Vector Machine (SVM) could be an index of academic ability. In this study, the prediction model was established by the link between the score of a mock examination and the pass/fail results of the 101st national examination for pharmacists of sixth-year students enrolled in 2015. To confirm the validity of the prediction model, we compared the score of a mock examination of sixth-year students enrolled in 2016, the predicted pass/fail results of the 102nd national examination for pharmacists derived by the prediction model, with the actual pass/fail results. These results suggest that the prediction model could predict pass/fail results with relatively high precision and be useful as an index of the transition of academic ability.