A Multistrategy Machine Learning Student Modeling for Intelligent System Based on Blackboard Architecture

• Main Authors: Yu Jung Hsieh, 謝育融
• Other Authors: Mu Jung Huang
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/39021315606180244861

碩士 === 國立彰化師範大學 === 資訊管理學系所 === 99 === This study tries to propose a blackboard approach that uses multistrategy machine learning student modeling techniques to learn for achieving the goals that are set on learning what the properties of the student's inconsistent behaviors are. These multistrategy machine learning student modeling techniques include inductive reasoning (similarity-based learning), deductive reasoning (explanation-based learning), and analogical reasoning (case-based reasoning). According to the properties of the student's inconsistent behaviors, the ITS (intelligent tutoring system) may then adopt appropriate methods, such as intensifying teaching and practice on the topic where the student's inconsistent behavior has happened, to prevent the student's inconsistent behaviors. The instructional component in the ITS also can refer the properties of the student's inconsistent behaviors to select more appropriate teaching strategies to teach the student. This research set the learning object on a single student. After accumulating these inferences from a group of students, we might also learn from them about what kinds of students are easy to have inconsistent behaviors or what conditions most students are easy to have inconsistent behaviors.