Please use this identifier to cite or link to this item:
標題: 類神經模糊控制在個人化線上評量系統的應用
The Application of Personalized On-line Evaluation System Based on Neuro-Fuzzy Control
作者: 黃義立
Huang, Yi-Li
關鍵字: neuro-fuzzy control;類神經模糊控制;learning model;adaptive difficulty;學習能力模型;適性難易度
出版社: 電機工程學系
本研究的目的乃利用「Neuro-Fuzzy Control」與「教材與學習的特性」實作「個人化線上評量系統」,其中包含使用Inverse learning設計方法與Carroll提出的學習模型,使得能夠提供給上線者適合其當時學習狀況的難易度評量,而且一份適性難易度評量各章節題數的決定是依據上線者之前評量的作答情形而有不同比例,以達到在較短時間各章節能整體性平衡的學習,至於題庫內各題目難易度也藉由累計所有上線者的作答情形作更新,以介於0到1間的連續數表示。最後比較適性難易度與隨機難易度在學習上呈現的差異,並討論整個系統中「教師Plant策略模型」與「預期學習能力模型」對適性難易度評量的影響。

The purpose of this thesis is to implement a personalized on-line evaluation system based on “neuro-fuzzy control” and the “teaching-learning characteristics”. We use inverse learning method and the learning model proposed by Carroll to design the system. It can provide an exam whose difficulty is adaptive to the progress of on-line user. The system chooses questions based on user's previous performance and pick more questions in the sections or chapters where the user performed poorly in the past. Thereby, the user can learn the materials more quickly without wasting time on the part they already understand. Instead of using a discrete difficulty measure for each question, the difficulty measure in database is also updated every time someone takes the exam and it is expressed by continuous numbers between 0 and 1. Finally, we will discuss the differences between the adaptive evaluation we proposed and the random evaluation which is currently in wide use. We will also discuss the influence of the “teacher''s plant strategy model” and the “student's expected learning attitude model” on the adaptive difficulty evaluation.
Appears in Collections:電機工程學系所

Show full item record
TAIR Related Article

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.