We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
NEURAL NETWORK BASED APPROACH FOR PREDICTING LEARNING EFFECT IN DESIGN STUDENTS.
- Authors
Ying-Jye Lee
- Abstract
This study examines a neural network based approach for predicting learning effect in design students. This investigation takes the percentile-grades of all courses taken by first-year design students, including Introduction to Industrial Design, Engineering Graphics, Basic Design, Technical Drawing, Chinese, English and Mathematics, and uses these percentile-grades as the input of the back-propagation neural network (BPNN). Additionally, the percentile-grades of professional core courses at the upperclassman level, including Form Design, Product Design and Thesis Project, are used as the output of the BPNN. Analytical results demonstrate that the BPNN model offers relatively accurate predictions for the student learning effect in the professional core courses, especially Thesis Project, with average accuracy of 93.54%. These analytical results indicate that BPNN is a suitable instrument for predicting the learning effect of design majors. Design instructors or student career consultants can use the BPNN model to identify individual students as having particular potential in design, and thus can tailor their teaching strategies, and provide additional guidance and assistance for individual students.
- Subjects
ARTIFICIAL neural networks; ARTIFICIAL intelligence; PRODUCT design; INDUSTRIAL design; ENGINEERING graphics education; INDUSTRIAL designers; TEACHING; COLLEGE students; LEARNING
- Publication
International Journal of Organizational Innovation, 2010, Vol 2, Issue 3, p250
- ISSN
1943-1813
- Publication type
Article