Build a predictive system for assisting the middle school graduates to choose their suitable high schools by using Fuzzy Artificial Neural Network

• Main Authors: Ke Chun Wei, 柯俊偉
• Other Authors: Yeh Dowming
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/94018882754185131621

碩士 === 國立高雄師範大學 === 資訊教育研究所 === 95 === The idea of the research was to build a predictive module for assisting the middle school graduates to choose their more suitable or odds-on senior high schools while taking personal interests, parents’ expectations and senior high school reputations into account. The study consisted of two parts, one applied the advantages of Back Propagation Network and the other one adopted the fuzzy logic theory. Technically speaking, the accumulating experiences and analyzing collected results of Back Propagation Network could build a predictive system which provided the average score through mock exam results as the predictive score of Basic Competence Test for each graduate. With the fuzzy logic theory, the degrees of conceptual distinction of matriculating probability could be transformed into the mathematical descriptions; for instances,“have a chance”,”it is feasible”,”it must” and “it is odds-on” could be measured then defined based on the required enrolled scores of different senior high schools. Finally, the linguistic concepts would be directly represented after inputting the obtained predictive grades or the names of senior high schools.