A machine learning approach for intelligent tutoring systems
Hussian F. Sindi · International Conference on Systems · 2005
The combination of artificial intelligence (AI) technology and education results in different products of intelligent educational software for all tasks and domains. With AI methods new generation of intelligent tutoring systems (ITS) and intelligent authoring tools can be created. Machine Learning Techniques provide a variety of methodologies and theories about reasoning, inference and learning. Hypotheses derived from AI theories can inform curriculum, pedagogy, and potential roles for computers in education. Machine learning based ITSs can adjust its tutorial to the student's knowledge, experience, strengths, and weaknesses. It may even be able to carry on a natural language dialogue. In addition, automatic generation of exercises and tests is an important feature of ITS. ITSs are complex to build, complex to maintain and face the knowledge-acquisition difficulty. This paper presents an overview of the machine learning techniques in intelligent tutoring systems. Also, the paper presents a proposed architecture of ITS based on case based reasoning paradigm. Additionally the work has been implemented in the prototype in biology domain.