Developing predictive decision support system for nursing licensure examination results using decision tree growing methods

Abstract

In data mining, the decision tree learning has seen to be one of the valuable classifying techniques that were found very useful to a wide range of data primarily to academic data. Monitoring of students’ academic performance is one of the main topics of educational data mining especially if a given course has a licensure examination¬. In the Philippines, licensure examination is one serious aspect in the field of education. However, still very few researches have been carried out concerning predictions of licensure examination performance. Most researches focus on academic retentions and developing dropout models. With this, the study is aimed to extract predictive model using Decision Tree Algorithm. The study aims to compare the performance of the decision tree growing methods in predicting student licensure examination using Confusion Matrix Test. These methods consist of Chisquared Automatic Interaction Detection (CHAID), Classification and Regression Trees (CRT) and Quick and Unbiased Efficient Statistical Tree (QUEST). The extracted rule sets from the algorithm will be embedded in a decision support system that could early predict and identify who will fail in the nursing licensure examination so proper academic support programs can be formulated.