Research On DM-based Personalized Portal

Portal can integrates all kinds of Mismanagement information system) and information into together. Through a portal, employees can share enterprise information and manage them orderly. After that, "information island" will be disappeared and the informationization level will be improved.From Internet portal to EIP(enterprise information portal) and EKP(enterprise knowledgement portal), the portal technique has been more and more mature, and there are more and more portals available. To develop e-commerce, enterprise must attract more customers through the portal. It's not an easy thing to stand out from many portals and attract more customers. So more and more people focus on the develop of personalized portal, and personalized portal has become one of the developing direction of portal.The concepts, types and significances of portal are introduced in this paper. Then we get two most important developing directions of portal—standardization and personalization. After that, some patterns of data mining are discussed, and their applications in personalized portal have been predicted. Later on, the algorithm of sequence pattern mining is discussed in detail, and an algorithm to developing personalized portals is put forward. In the end, a simple personalized portal has been developed and realized based on this algorithm.In this paper, we get the significances of personalized portal through the analysis of portal technique. Meanwhile, the feasibility of using data mining in personalized portal is proved through the study of data mining.There are three new ideas in this paper. The most important developing direct of personalized portal—intelligent personalized portal—has been put forward. Through the rebuild of sequence pattern algorithm, we get a new algorithm to develop personalized portal. A simple intelligent personalized portal has been developed and realized based on this new algorithm.