A Personalized CAI Courseware System

Abstract

The network-based CAI (Computer Assisted Instruction) courseware system,Compared with the traditional one, has its advantages and disadvantages. After give an introduction to the ongoing development of distance learning and E-learning. This paper combined a hierarchical Level-Generate algorithm with the Data Mining theory, to provide the students with a personalized courseware system by building a CAI courseware tree. With courseware tree, the system can recommend useful materials to each student based on the performance of individual student. If the system computes the personality of each student, the complexity is higher and hard to implement. Thus, mining on students is necessary. Using Hamming distance theory, students were clustered into different groups based on the courseware they accessed. Computer Assisted Instruction, abbreviated as CAI, is now widely used in all kinds of educational applications, especially the distant learning and network-based tutor systems, while the network technology is prevailed all over the world. However, the distance learning and most applications of the courseware are restricted to traditional classroom and only used locally. The Internet profoundly changed the way people communicate and becomes available everywhere. The traditional teaching system is now challenged by the Internet and many researchers exert great effort on developing the network based learning systems to assistant the traditional classroom teaching. The network based CAI courseware, with its vivid appearance, interchangeable ability and the breakthrough of time and space, declares an evaluation to the traditional teaching model. During the past several years, researchers provide the educators with some tools that auxiliary the traditional teaching systems. In (1), C. Romero and S. Ventura survey the application of data mining to traditional educational systems, especially web-based well-known learning content management systems, and adaptive and intelligent web-based educational systems. The author firstly compared the difference between e-commerce and e-learning. The authors also distinguished between three different types of web-based education systems: particular web-based courses, well-known learning content management systems, and adaptive and intelligent web-based educational systems. Then, four kinds of data mining techniques were presented: statistics and visualization; clustering, classification and outlier detection; association rule mining and pattern mining; and text mining.