Applying fuzzy neural network construct web-based system to infer learning effect

• Main Author: 李孟柔
• Other Authors: 劉遠楨
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/61169658889842673092

碩士 === 國立台北師範學院 === 數理教育研究所 === 91 === Generally, the traditional teaching pedagogy depends on whether a student “understands or not” to evaluate his/her learning; however, there is a huge gray area between the two poles of understanding. Therefore, giving an answer that does not conform to the correct one in a learning evaluation does not mean that the student has no understanding of the concept. In this study, fuzzy theory and artificial neural network techniques are combined, with the convenience of the Internet, to construct an inference system to evaluate a student’s learning. Fuzzy theory is employed to put the data of teachers’ questionnaires together, form the weight matrix between concepts and concepts and the weight matrix between concepts and tests, combine a three-layered artificial neural network structure. The structure would infer a student’s learning progress according to his/her test data and combines it with fuzzy theory to build a fuzzy medium and to provide an oral description of evaluation and advices. This system is individual, thorough, compatible, and for distance learning.