Improve Contextual IoT service discovery with semantic models

The Internet of things (IoT) is an ecosystem of smart connected devices that exchange data over a communication network. By integrating these devices into different vertical applications, the IoT has the potential to have a major impact on both the economy and society. However, the plethora of heterogeneous devices with varying ways of describing the information raise interoperability issues. In this context, the development of appropriate service discovery mechanisms enriched with semantic capabilities for understanding and processing context information is a key feature for turning raw data into useful knowledge and ensuring interoperability among different devices and applications. In previous work, we focused on surpassing the IoT semantics barriers while exploring novel networking approaches. To this end, we proposed a service discovery mechanism, realised on top of Named Data Networking (NDN), that relied on a semantic matching mechanism for achieving a flexible discovery process. Since the initial work, several improvements were made to the semantic similarity model at the basis of the semantic matching algorithm. This work replicates the scenario proposed on the former contribution and assesses the impact of the improved semantic model. Results show that while the previous semantic model achieves a mean Average Precision of 0.29, the best performing current solution achieves 0.68.