WSDL Retrieval for Web Services Based on Hybrid SLVM

Recently, two operable WSDL retrieval approaches, bipartite-graph matching and KbSM, were developed for Web service discovery. But their models and similarity metrics of WSDL ignore some term or semantic feature, and involve formal method problem of representation or difficulty of parameter verification. SLVM approaches depend on statistical term measures to implement XML document representation. Consequently, they ignore the lexical semantics and the relevant mutual information, leading to dataset analysis errors. This work proposed a service retrieval method, hybrid SLVM of WSDL, to address above problems in feature extraction. This method constructed a lexical semantic spectrum using WordNet for characterizing the lexical semantics, and built a special term spectrum based on TF-IDF. Then, feature matrix for WSDL representation was built on the hybrid SLVM. Applying to NWKNN algorithm, on dataset OWLS-TC-v2, our method achieves better F1 measure and query precisions than bipartite-graph matching and KbSM.