Study of Query Interface Matching Based on Singular Value Decomposition

Deep Web query interface matching is one of the key technologies to realize rapid Deep Web data query. On the study of query interface representation, feature extraction and weight calculation, the method based on singular value decomposition is presented. It transforms the query interface set into a feature-interface matrix, selects the final classification features and creates a feature vector matrix. In order to reduce the computational overhead of similarity computing between the different query interfaces, the singular value decomposition is adopted to truncate the vector matrix, which will build a low rank matrix to approximate the original feature vector matrix. This method realizes the fast matching between the original query interface and the target query interfaces, improving the efficiency of Deep Web online database query.