A Novel Method on Incremental Information Acquisition for Deep Web

Abstract

Deep Web is autonomous, independently updating, and its data are always in a state of frequent update. However, the user always hopes to obtain the newest content in the current Web database. Different from previous research, this paper wants to emphasize the importance of updating frequency in the study of Deep Web information acquisition. And, an approach on incremental information acquisition based on logical reinforcement learning has been proposed. Then, we find in our research that under the same condition of constraint resources, the novel approach can improve the freshness of data, discovery efficiency of new data and the service quality of Deep Web information integration.