Semantic clustering analysis for web service discovery and recognition in Internet of Things

Abstract

Today, semantic web services are rapidly evolving and updating. The discovery of semantic web services is an important concept for the comprehensiveness of individual web services in creating new intelligent systems that meet the complex needs of users and is an important technology in the domain of web services. One of its main goals is to reuse existing web services and combine them in a process that has attracted a lot of attention from different communities like the Internet of Things (IoT). Currently, the discovery of web services in the most common category includes four main methods and a set of sub-methods. Most semantic web service discovery methods are done using semantic descriptions of web services using ontology based on existing pattern recognition approaches. In this research, a new approach is presented that in the first step, the web services description language (WSDL) is scanned and the infrastructure is examined. In the second step, by adding the technique of extracting background information that can be received from the WSDL, the field limitations and existing patterns are considered and detected by semantic spacing on the discovering web services. Also, web services whose parameters contain synonymous synonyms, irregular composite fragments, and similar abbreviations with high accuracy in a cluster contract. The proposed approach is based on a semantic pattern recognition using data mining and finally, the output of the proposed method is single and combined services that have high accuracy and speed of the proposed algorithm in web service discussions.