Akademik Veri Yönetim Sistemi
International conference on innovative engineering applications, Sivas, Türkiye, 20 - 22 Eylül 2018, ss.988-996
Wireless Sensor Networks (WSN’s) has become a vital part of electromechanical systems in many different applications. These systems collect a huge mass of data in a short time period. However, these data still represented by classical methods in many sensor applications. These classical methods cover relational database, XML base files, or any syntax-level integration format. However, there is a need for more effective methods for better understanding and analysis of such huge sensor data. In this study, an ontology-based system is created for the data obtained from the sensors in order to more meaningful represent the sensor data. This system was named as OReHSeD (Ontological Representation of Heterogeneous SEnsor Data). Thus, semantic integration was provided successfully with sensor data. The proposed system has been implemented in greenhouse monitoring application as a real-world use case. The results of the application show that the OReHSeD system can successfully represent raw sensor data as meaningful semantic data. In future work, sensor data can be clustered or classified using machine learning algorithms in order to predict the unforeseen situation in the near future. This process may be useful for future estimation of the observed feature. In this way, some measures can be taken to prevent unwanted situations and natural disasters such as earthquakes, forest fires, floods.