Prediction Models for Licensure Examination Performance using Data Mining Classifiers for Online Test and Decision Support System

This study focused on two main points: the generation of licensure examination performance prediction models; and the development of a Decision Support System. In this study, data mining classifiers were used to generate the models using WEKA (Waikato Environment for Knowledge Analysis). These models were integrated into the Decision Support System as default models to support decision making as far as appropriate interventions during review sessions are concerned. The system developed mainly involves the repeated generation of MR models for performance prediction and also provides a Mock Board Exam for the reviewees to take. From the models generated, it is established that the General Weighted Average of the reviewees in their General Education subjects, the result of the Mock Board Exam and the instance when the reviewee is conducting a self-review are good predictors of the licensure examination performance. Further, it is concluded that the General Weighted Average of the reviewees in their Major or Content courses is the best predictor of licensure examination performance. Based from the evaluation results of the system, the system satisfied its implied functions and is efficient, usable, reliable and portable. Hence, it can already be used not as a substitute to the face-to- face review sessions but to enhance the reviewees’ licensure examination review and allow initial identification of those who are likely to have difficulty in passing the licensure examination, therefore providing sufficient time and opportunities for appropriate interventions.

Keywords – Performance Prediction Models, Data Mining, Decision Support System