An Efficient Ontology‐Based Semantic Interoperability Using MSGO‐RNN in Cloud Computing.

Semantic interoperability (SI) is defined as the capability of interpreting the nature of the information exchanged inside cloud computing (CC). For SI, ontology is selected as a solution. A hierarchical structure is offered by an ontology that comprises semantic relations between the application and the cloud server (CS). Even though different methods are introduced, ontology centered on semantic relation between the application and CS is not yet attained for achieving accurate interoperability. An ontology‐centered SI in CC utilizing a modified Shell Game optimization‐centered recurrent neural network (MSGO‐RNN) is presented in this work. Here, accessing the data from the data warehouse is initiated that experiences different standardization functions for improving the data's quality, namely, data extraction, data cleaning, data transformation, data loading, and refreshing. Then, the data are passed via the semantic layer (SL), which offers a constant way of interpreting data. Then, for enhancing the accuracy rate of interoperability along with offering a low computational time, the relevant feature is chosen utilizing genetic crossover mutation‐centered improved fertile field (GCM‐IFF) optimization. Ontology is created utilizing the Protégé tool centered on the data, and respective manipulation is executed utilizing HermiT ontology web language (OWL) reasoner. Lastly, to attain cloud interoperability, the features are trained and tested by employing the MSGO‐RNN. The experimental outcome exhibits that when compared with the top‐notch methods, an accuracy of 96.32% is acquired by the framework for choosing the CS, and less time of 22567 ms is acquired by the framework for training. [ABSTRACT FROM AUTHOR]