Adaptive Web-Based Educational System using Neural Networks in EFL Course

A new framework for Web-based intelligent educational system (WBIES) is developed, incorporating bidirectional Associative Memory (BAM) neural network to represent the knowledge domain and a Fuzzy BAM to evaluate the student's knowledge status related to each studied concepts of the course. The domain model is built using a hierarchical BAM of learning objectives, concepts and educational units. The proposed framework allows dynamic generation of the hypermedia pages depending on six levels of student's knowledge status which represent the output of FBAM. Experimental results show that the new system generally exhibits faster performance and better adaptivity for content presentation and page hyperlinks. This approach is experienced in English as a Foreign Language (EFL) course for Arabic beginner's students.