Research On Database Choice And Query Conversion For Deep Web

With the rapid development of World Wide Web, Deep Web contains a mass of information available for access, and it is still in rapid growth. Through on-line query interface access their back-end database of the Web the information is obtained. Despite the wealth of information is hidden in the Deep Web, Deep Web data is as a result of heterogeneous and dynamic, to effectively make use of this information is a very challenging task.At present, the research of Deep Web in this area has made the system architecture of Deep Web data integration, which is divided into two major modules: generating integrated query interface module and query processing module on integrated query interface. In this paper, the various components of the Deep Web data integration system architecture are discussed, and the choice of Web database and converting query in the query processing module are described in detail and researched in depth.As the continuous rapid growth of the number of Deep Web database in each area, if only briefly the query on the integrated query interface is converted into the queries of the each Web database in this area, to visit the each database in this area will be a lot of the cost, and a majority of Web databases in the field have redundant information, the results of these queries will have a large amount of redundant information. Based on the above issues, this paper presents based on TOP-N for the choice of Deep Web database, and improve distance calculation of irrelevant degrees and the choice method of Deep Web database, making the choice of Deep Web database to reduce the waste of resources, and improving performance of the choice of query.The query conversion is also an important part of the Deep Web data integration system architecture, which will be responsible for converting the user's query into the queries on all local target query interface of accessing Web database. With the rapid development of Deep Web data sources, the data sources in each thematic field are constantly updated or increased, the local target query interface in each area also constantly change dynamically, so changing users submitting requests for query into the queries on all local target query interface of accessing Web database is very difficult. Based on this issue, modeling the query conversion mechanism is deeply researched in the paper, and the paper gives solutions to improve the accuracy and performance of the query conversion.