An Ontology-Aware Model-Driven Approach for Service-Oriented Application Development: A Stepwise Refinement Manner

Ontologies are attracting increasing attention in software engineering research due to their ability to precisely model the semantic aspects of systems. Enriching software system models using ontology principles, especially in the process of developing interactive systems like Service-Oriented Architecture (SOA), can lead to the automated production of high-quality codes. Additionally, ontology-aware specification of the service from the abstract to the concrete level leads to early precise extraction of the service required by the users. This paper introduces a model-driven ontology-aware service development process to reduce the burden of code generation. This integrated approach utilizes a stepwise refinement methodology facilitated by a novel refinement algorithm to automate SOA development. The effectiveness of the approach is evaluated using three proposed parameters which examine the characteristics of the refined model in each refining step through some practical SOA case studies. Finally, we calculate the recall, precision, F-measure, accuracy, and also time analysis of discovered querying services for various scenarios in AWS and Netflix before and after applying ontology. The results show an average of 17% improvement in these metrics after applying ontology.