E-learning Environment, User Interest Model With Applications

With the development of World Wide Web(WWW) technology, Internet has become one of the most importent information and knowledge sources. E-learning has achieved great success both in research and applications in recent years, and provided users with unprecedented abundance of learning materials and great flexible learning ways. However, partly because of technical limitation, most E-learning system still focus on the web site now, which leads to the case that students don't know where the information and knowledge what they enjoyed and required is. Learners would like the E-learning systems can cater to their interests through adjusting web pages, so that everyone can feel the system offer services for him(or her) only. Consequently, designing and constructing a personalized E-learning environment has attracted much attention of researchers from the area of web mining, educational technology and artificial intelligence etc.In this paper, we focuses on investigating the user interest modeling technology which is one of the key problems of personalization. Our purpose is to construct a sound and effective mathematic model of E-learning to describe and analyze the latency knowledge in complex user behavior, that is to mining and analyze learners learning interests, which can further recommend the learning resources to users.Firstly, the key technologies of personalization and the application of personalized technology in E-learning are studied. And some of related knowledge is introduced, which include web mining, China E-learning Technology Standardization(CELTS), user modeling etc. Secondly, after investigate the technology of user interest mining, a user interest mining model is proposed. The model is based on user viewed contents and the character of E-learning. We studied and analyzed the process of this model, including date gathering, pre-processing, feature extraction and feature weight, document model based on Vector Space Model, user interest model and so on. As the two important aspects, text categorization and user interest model is introduced especially. After that, user interest modeling and user profile building are discussed, and some concepts are given: user interest degree and two layers structure of user interest model, which then be expressed in XML. Finally, a user interest mining prototype is designed and implemented, it can recommend the learning resources to learners through web page's template based on user profile.Last but not least, the later research is improving the algorithm, recommends exact...