Application of Fuzzy Logic to Diagnostic Testing In an E-Learning Environment

Chew, Meng Sang and Makati, Tanuj (2011) Application of Fuzzy Logic to Diagnostic Testing In an E-Learning Environment. INTI Journal: Special Issue on Teaching and Learning. pp. 1-10. ISSN 1675-0284

Preview

Text 2011_1.pdf Download (1MB) | Preview

Abstract

In an e-Learning environment, before a student moves on to the next module a diagnostic test Is given from which a grade is provided. However, it is far more useful to the student 1f he is given some Indication as to his actual capability In relation to the instructor’s expectations. A letter grade or a numerical score (except at the extreme ends of the range) while desirable, does not provide sufficient insight into how the student lairs. What is needed is feedback Lo the student that helps him in his continual Improvement in the subject matter. It is the myriad of variables such as, the student’s speed in working out the problems, guessing, and the fraction of correctly answered problems. A careful consideration of these new variables gives this investigation a fresh outlook to the problem of diagnosis that has not been previously addressed. This article presents an application of Fuzzy Logic to diagnose a student’s mastery of a narrow sliver of subject matter that he has studied. Recommendations are then given as to how he should proceed with his studies. The power of this approach lies in the ability of the computer administrating the diagnostic test to track a student’s problem-solving speed for each and every problem that is attempted. Interestingly it is also possible to determine ¡f (he student made guesses Instead of actually working out the correct answers to the problems. For diagnostic purposes, a correct answer from exercising a guess should not count towards the number of correct answers but as an unanswered question, Such a capability is not available for in-class tests because the actual answering speed for each question Is not tracked, nor is it possible to determine if a correctly answered problem comes from guessing at the answer. This article describes the workings of a Fuzzy Logic diagnostic program and how it addresses all these interesting variables and incorporates them into a useful tool for assisting e-Iearning students in determining their mastery of the subject matter.

Actions (login required)

View Item