A Classification model of cognitive learning objectives based on Bloom’s taxonomy using Naïve Bayes classifier

Abstract

Specification of course learning objectives in a curriculum is aimed at achieving the desired learning outcomes. The purpose of such a specification is to have some standardization of the teaching-learning process to achieve the learning objectives. A clear course learning objective improves the interaction between the instructor and the learners and helps the learner to know what skills and knowledge the learning outcome, is expected after completion of the course. This independent study aimed to propose a classification model of course learning objectives into the cognitive level based on Bloom’s taxonomy. As instructors sometimes set each own course learning objectives with ambiguous verbs. The curriculum supervisors then faced the problems in assessing the course learning objectives which is the main part of curriculum development. To solve the problem, this study proposed a classification model of course learning objectives based on the revised Bloom’s taxonomy using the Naïve Bayes classifier. Our proposed methodology is based on keywords extraction using a rule-based approach, machine learning, and data mining to automate the entire process. We are currently using data sets of learning objectives of bachelor’s degree level in the academic year 2019 of Faculty of Science, Chiang Mai University for testing purposes. This study used 3 classifiers such as Naïve Bayes, K-NN, and Decision tree in comparing the average classification accuracy for 10-fold cross-validation. The results showed that the Naïve Bayes classifier had the highest accuracy at 83.51%. The classification model could be applied to various fields to improve curriculum designing, curriculum evaluation, and more.