This paper presented personalized web agent that constructs user profile which is consisted of user preferences on the web and recommends his/her concrete information to the user. The monitor agent make user describe his/her preference directly and create preference document database, finally performs several keyword extraction to increase the accuracy about the DB. The learning agent transform the extracted keywords into user profile that could be confirmed and edited by the user. In this paper, we describe the several keyword weighting and inductive machine learning modules in detail. Finally, we describe the adaptive web retrieval and push agents that perform adaptive service to user.
展开▼
