A WEB-BASED TOOL FOR TEST BANK QUESTIONS SELECTION BASED ON CLASSIFICATION MODEL

• Main Author: Koh, Kia Hui
• Format: Final Year Project Report
• Language: English; English
• Published: Universiti Malaysia Sarawak, (UNIMAS) 2015
• Subjects: QA76 Computer software
• Online Access: http://ir.unimas.my/id/eprint/40200/1/KOH%20KIA%20HUI%20%2824%20pgs%29.pdf; http://ir.unimas.my/id/eprint/40200/4/KOH%20KIA%20HUI%20%28fulltext%29.pdf; http://ir.unimas.my/id/eprint/40200/

Examination questions selection is very important for lecturers to design the examination papers. As choosing the appropriate questions might affect the passing rate of students who participated in the examination. However, there is no suitable question generator tool that could help lecturers to select the examination questions based on item analysis metrics. By using the data mining techniques, we can discover the knowledge from the educational domain data. This valuable knowledge can apply into a system to predict and select examination questions. Therefore, this project aims to develop a web-based question generator tool by using predictive data mining techniques. The classification model in this tool is applying the concept of data mining classification method, known as Naive Bayes Classification to classify the questions in datasets. The question selection tool is able to generate a list of exam question from the database of classified exam question based on pre-determined passing rate. A simple calculation is used to calculate the passing rate that is determined by the user. Hence, the question generator tool should be capable to help lecturers in selecting examination questions easily and efficiently.