Generating the Semantic Containers for the Query Interfaces of Deep Web

With the myriad emergence of the online Web Database, the Web is divided into a two layers information platform which is composed of surface Web and deep Web. The huge information hidden in the deep Web is higher in the quality than the surface Web. This interests the researchers of the traditional database community. In order to adapt the Web community, the idea of information fusion is further transformed. However, the nature of the Web is great different from the database in structure, principle and function. Query reformulation is one of the technical challenges. The only accessible method to get the information in the deep Web is through the query interfaces provided by the Website of deep Web. There are many semantic restrictions between the attributes in the query interface. Some attributes are co-occurred and the others are exclusive sometimes. To generate a valid query, we have to reconcile the key attributes and their semantic relations. To address the problem, a method based on the HTML codes is presented in this paper. Different kinds of semantic containers can be got through analyzing the codes of the query interface. The experiment is carried out on the famous dataset of deep Web to show the feasibility of the method.