The Study Of Deep Web Data Integration System Design And Application

Web contains massive valuable information. The information, according to the“hidden” depth, can be divided into sub-categories: Surface Web and the Deep Web.The former refers to that a traditional search engine can search the set of static pages,while the latter opposite. Deep Web mainly refers to a collection of dynamicallygenerated pages from the Web database (WDB) by users submission query on the queryinterface of WDB. According to the statistics published by industry authorities, theinformation contained in the Deep Web is about500times that in the Surface Web.Besides, the information in the Deep Web seems more professional since they can befocused on a specific domain and structured data. Thus, it turns to be fairly importantand urgent for us to mine the useful information in Deep Web.An efficient method for achieve the rich and valuable information “hidden” in theDeep Web is by the Deep Web data integration systems. Based on the basis of extensivestudy to the key issues of the Deep Web data integration system: the Deep Web dataintegration system architecture, Web information extraction techniques or algorithms,query interface model extraction algorithm and the query interface integration, thispaper introduces a Deep Web data integration system framework and an improved webinformation extraction MDR (Mining Data Records) algorithm. Meanwhile this paperpresents a query interface integration algorithm based on ontology and Java technology.By analyzing the actual requirements, the Deep Web data integration technology hasbeen introduced into the design and development of practical projects, which achievedgood social and economic benefits.