Building Semantic-Based Applications in IoT Environments With IoTO++

Abstract

Nowadays, the Internet of Things (IoT) integrates connectivity into devices to improve processes and optimize resources in various scenarios (e.g., hospitals, factories, and homes). However, the diversity of technologies and devices makes interoperability a challenge. For this reason, ontologies have emerged to model knowledge and to establish guidelines for proper communication. As the number of devices and the volume of data increase, additional considerations, such as data privacy, security, environmental awareness, and ethics, become essential to ensure appropriate data protection and efficiency. To address these issues, we demonstrate the applicability of the IoTO++ ontology, which aims to represent not only the core components of the system (such as sensors and actuators) but also the data they produce. A museum simulation was created in Node-RED that models an IoT ecosystem with five types of sensors and an access control system for visitors and personnel. This application was validated in order to evaluate correctness and quality. To this end, correctness was assessed using the Pellet reasoner, while quality was evaluated through tailored questionnaires. These assessments were successfully addressed using both description logic representations and SPARQL queries, thereby demonstrating the ontology’s expressiveness and its alignment with domain requirements. This dual perspective further strengthens the ontology’s credibility and readiness for deployment in diverse IoT environments. Performance analysis, averaged over ten runs on a dataset of 100 sensor observations and 1000 triples, produced query execution times between 0.0017 and 0.0268 s ( $\approx$ 0.0061 s average). Furthermore, scalability tests with 10–10 000 observations showed stable loading times ranging from 0.58 to 0.62 s, averaging 0.5931 s. These results demonstrate the system’s efficiency and consistent performance for real-time IoT applications as datasets scale.