Research And Realization Of ICAI System Based On Web

The Web Based Intelligent Computer-Assisted Instruction system is in the last few years along with the rapid development in network technique and artificial intelligence but newly arisen research topics. Meanwhile, it is also a complex system involved with computer science , pedagogy, psychology, cognitive science and behavior science. It aims to provide the computer system with wit, so the computer system can take the place of human teacher to instruct the student in some degree. Using the system, the learner may break through restrict of traditional education resources and the education method, being free from the conditional stipulations, such as time and space etc., applying for the individual study according to the own knowledge level and time arrangement.This article embarks from the actual demand of the Intelligent Computer-Assisted Instruction system, has analyzed system model, gives a detailed elaboration on the main function of the system, and creats an ICAI system based-on Web. Here, a course is divided into separate knowledge points, of which, the interrelation is expressed in knowledge trees. According to the features of knowledge points, system adopts production rule representing method, frame-representing method, the semantic web representing method and object-oriented representing method to explain the knowledge information of the system. Meanwhile, under the cognitive theory instruction, we construct the cognitive student model in accordance with students' study progress, and realize this model by step-by-step approach method. Thus promotes the system individualized teaching ability and promises to do better to teach the students in accordance with their aptitude.Finally, through to the key research of exam paper generating algorithm—core question of Intelligent Exam System, on the foundation of traditional genetic algorithm, proposes the auto generating exam paper strategy based on integer coding and adaptive genetic algorithm, which increases the efficiency and speed of the generation of exam paper greatly. And evaluates the exam paper made by genetic algorithm with the fuzzy comprehensive evaluation way as judged by four factors: paper range, paper difficulty, the amount of question as well as the variety of question type, and gathering of creditability of experts. In this foundation, uses B/S pattern, utilizes JSP technology and the MySQL database, and has realized an Intelligent Exam System.The application of the proposed system is supposed to efficiently improve students' original cognitive structure, laying stress upon the development of their ability of analyzing and solving problems, provides more intellectualized, more convenient, highly effective network-learning environment for more and more learners. This system is also of reference value to other ICAI researchers and developers.