Abstract
Sensors have been growingly used in a variety of applications. The lack of semantic information of obtained sensor data will bring about the heterogeneity problem of sensor data in semantic, schema, and syntax levels. To solve the heterogeneity problem of sensor data, it is necessary to carry out the sensor ontology matching process to determine correspondences among heterogeneous sensor concepts. In this paper, we propose a Siamese Neural Network based Ontology Matching technique (SNN-OM) to align the sensor ontologies, which does not require the utilization of reference alignment to train the network model. In particular, a representative concepts extraction method is presented to enhance the model’s performance and reduce the time of the training process, and an alignment refining method is proposed to enhance the alignments’ quality by removing the logically conflict correspondences. The experimental results show that SNN-OM is capable of efficiently determining high-quality sensor ontology alignments.
Highlights
Over the past decades, sensors have been growingly used in a variety of applications, e.g., medical science, space observation, wildfire detection, traffic management, and weather forecasting (Yawut & Kilaso, 2011; Doolin & Sitar, 2005; Topol, Steinhubl & Torkamani, 2015; Chen, 2018; Sheth, Henson & Sahoo, 2008; Gravina et al, 2017; Du et al, 2020; Chu et al, 2020)
For the efficiency of ontology matching, several machine learning (ML) algorithms have been studied by them, i.e., decision tree (DT), AdaBoost, K- Nearest Neighbor (KNN), and support vector machine (SVM), and the results shows that the combination of DT and AdaBoost classifiers outperforms others
An alignment refining method is proposed to enhance the quality of the alignment, which makes use of the sensor ontology’s concept hierarchy to remove the logically conflict correspondences
Summary
Sensors have been growingly used in a variety of applications, e.g., medical science, space observation, wildfire detection, traffic management, and weather forecasting (Yawut & Kilaso, 2011; Doolin & Sitar, 2005; Topol, Steinhubl & Torkamani, 2015; Chen, 2018; Sheth, Henson & Sahoo, 2008; Gravina et al, 2017; Du et al, 2020; Chu et al, 2020). As the complex nature of sensor ontology matching process, recently, the neural network (NN) becomes the popular technique to align the sensor ontologies (Khoudja, Fareh & Bouarfa, 2018b; Khoudja, Fareh & Bouarfa, 2018a; Bento, Zouaq & Gagnon, 2020; Jiang & Xue, 2020; Iyer, Agarwal & Kumar, 2020). The existing NN-based OMTs require the utilization of reference alignment, which is unavailable in the real-world matching task Their model training process needs long runtime, which hampers their applications. To overcome these drawbacks, this work proposes a Siamese neural network based ontology matching technique (SNN-OM) to effectively and efficiently align the sensor ontologies. ‘Siamese Neural Network Based Ontology Matching Technique’ gives the detail of SNN-OM, including the extraction of representative concepts, training of the model, matching process, and alignment refinement.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
