An approach to automatical semantic analysis of Web query interfaces

An increasing number of data emerging in Web brings difficulties to discovering and applying the information hiding on the deep Web. As the interaction interface between deep Web and outside, semantic of web query interface (WQI) needs to be analyzed in an accurate way. Most of the existing works merely deal with the single element. This paper proposes an automatic approach to analyze semantic of WQI by taking WQI as a whole of specific domain, using existed knowledge to construct match. And the algorithm verification is given by experiments of TEL-8 dataset in UIUC. Moreover, we present the optimization algorithm to improve the performance.