Hybrid cognitive model for semantic discovery and selection of services

Lack of (semi)automatic mechanisms for service classification in the Universal Description Discovery and Integration repositories and non utilization of explicit or implicit semantic information of a service during its publishing are the two major challenges in the area of web service discovery and selection. We propose a semantic model of human-machine collaboration for the classification, discovery and selection of web services that integrates the semantic as well as syntactic data of the web services to achieve the hybrid cognition. This proposed cognitive approach uses the principals from the machine learning, measures of semantic relatedness and information retrieval where the cognitive information from the WordNet based Omiotis measure of semantic relatedness is merged with the syntactic service profiles and further these semantically enriched service vectors are passed to the supervised learning algorithms to achieve the decision support for the discovery and selection of relevant services. Empirical evaluation of the proposed approach implemented on OWL-X data set has been presented and a comparison of two different supervised classifiers has been made.