Research On Deep Web Database Integration Technology

Compared with the Surface Web, Deep Web contains most information of the entire Web, and these data mainly from the back-end database, user can only accept it through the submission of a query requests. Because of these rich Deep Web resources can’t be directly retrieved by traditional search engines, it’s particularly important to integrate many different Deep Web database and acquire Deep Web data from different sites in unified form for user.At present,most Deep Web database integration technology is based on the field, the structure is similar for the Deep Web query interface mode information in the same fields. According to this principle, it integrates on the query interface of the same fields virtually. Based on the study and analysis the research status and the existed problems in integration methods of the Deep Web database integration technology at this stage, this paper proposes two kinds of Deep Web database integration technology. One is for none-real time query, the other is for real time query. This paper focus on the none-real time query. In the non-real time query Deep Web database integration technology, this paper works on the depth study of obtain data from the source, Deep Web database integration and query results synthesized these three areas, and in the last gives an example for the integration of two heterogeneous database using this method. In the real-time query Deep Web database integration technology, this paper introduces the classification and selection of Web data sources.Finally this paper analysis the advantages and disadvantages of the proposed two different integration schemes, and then compares each other with another Deep Web database integration technology which based on query interface. The results show that non-real time queries Deep Web database integration technology can solve the existing main problems in the database integration effectively, it also has high feasibility and research value.