Matching Sensor Ontologies with Neural Network

With the widespread adoption of sensors, many research efforts in recent years have focused on the wireless sensor networks. Since a wireless sensor network consists of a large number of sensors, its trusted communication become a hot research topic. Sensor ontology matching technology is able to solve the sensor information heterogeneity problem, which ensures the communication quality among different wireless sensor networks. In the matching process, different classes of similarity measures have different contributions in matching sensor ontologies. How to determine the optimal weights to aggregate multiple similarity measures to obtain high quality ontology alignment becomes a challenge in sensor ontology matching domain. To face this challenge, this work proposes a neural network-based sensor ontology matching technique. In particular, a single layer perceptron is used to aggregate multiple similarity measures and the neural model is trained with different training examples to obtain higher ontology matching accuracy. The experimental results show that the proposed approach is able to determine higher quality alignment results compared to other matchers under different domain knowledge such as bibliographic and real sensor ontologies.