Assessment of student competency for personalised online learning using objective distance. (2018)

• Record Type: Journal Article
• Authors: Chaichumpa, Sataworn; Temdee, Punnarumol
• Is Part Of: International journal of innovation and learning. Volume 23:Number 1(2018)
• Page Start: 19
• Page End: 36
• Publication Date: 2018
• Subjects: online learning -- personalised learning -- classification -- objective distance -- K-nearest neighbour -- KNN -- artificial neural network -- ANN -- decision tree; Knowledge management -- Periodicals; Learning -- Periodicals; Diffusion of innovations -- Periodicals; 006.33
• Languages: English
• ISSNs: 1471-8197

Abstract

Like traditional learning, online learning also requires effective personalised learning so that the appropriate feedback can be given individually for the students to achieve their goals. This paper proposes the objective distance which is the measurement representing the distance between current status of student's competency to the satisfied competency level required for accomplishing the entire course. This paper aims to study to what extent the proposed objective distance can be used for effective classification of student's competency comparing to raw score data and its combination. The experiments are conducted with two different online courses including computer skill and English language course having 55 and 111 students respectively. The students are classified with three classifiers including K-nearest neighbour, artificial neural network and decision tree into different classes accordingly to different competency levels. The classification results show that the proposed objective distance can be effectively used for competency classification of online students.