Abstract—The study of medical various institutions in
Thailand found that the success for teaching and learning by
means of Problem Based Learning (PBL) depends on several
factors including correct attitudes and cognitive learning.
Problem of teaching a small group of PBL is consistent with the
results of the audit were to evaluate the group's facilitator,
diverse and inaccurate. Especially, the grading of a medical case
study essay reports (Clinical case summaries). To solve such
problems, we proposed automatic medical case study essay
scoring for PBL of medical students. SVM with Genetic
Algorithms (GA-SVM) was used to assess the quality medical
case study essays written by medical students in the subject
matter of muscular systems and movement. The medical case
study essays written in response to a question were each
evaluated by facilitators and assigned a human score. In the
experiment, we used raw term frequency vectors of the essays
and their corresponding human scores to train the SVM while
GA was used for choosing the kernel function type and its
parameter values to find a proper solution to an optimization
and obtain the machine scores. The experimental results show
that the addition of GA-SVM technique improves scoring
performance.
Index Terms—Essay scoring, medical case study, SVM,
genetic algorithms (GA).
S. Yenaeng is with the Department of Computer Education, Faculty of
Education, Bansomdejchaopraya Rajabhat University, Bangkok, Thailand
(e-mail: modssk@gmail.com).
S. Saelee is with Department of Computer Education, Faculty of
Technical Education, King Mongkut’s University of Technology North
Bangkok, Bangkok, Thailand.
W. Samai is with the Faculty of Medicine, Prince of Songkla University,
Songkhla, Thailand.
Cite: S. Yenaeng, S. Saelee, and W. Samai, "Automatic Medical Case Study Essay Scoring by Support Vector Machine and Genetic Algorithms," International Journal of Information and Education Technology vol. 4, no. 2, pp. 132-137, 2014.