Cloud service mentor system using ambiguous connectives and refined user request handling methodology – a survey

Abstract

Cloud service selection under diverse nature of the cloud services is still an open issue. Unlike web services which are all selected based on WSDL description, cloud services cannot be selected as there is no standard representation of cloud services as most of the cloud services are described in natural language. In this paper we proposed a cloud service mentor system which is built upon semantic technologies and has three ontology structures for cloud services namely SaaS, PaaS, and IaaS ontology. This system parses the cloud service description documents through natural language handling method and major concept of the cloud service is identified. Based on the major concept of the cloud service the cloud service ontology is updated. The user request is processed using natural language handling methods and ambiguous connectives (AND, OR) are used to refine the request based on first order horn clause logic and DNF. Then the SPARQL queries are generated to retrieve the corresponding cloud services from ontology repository. The performance evaluation of the system shows that this system outperforms other methods and provides better results.