Ontology Annotations And Named Entity Combined Sensor Semantic Enhancement System

Internet of Things(IoT),according to the standard communication protocol,connects sensors,controllers,people and a series of things together,to achieve remote management control and intelligent through the relationships among things.A large number of sensors are deployed in IoT,the data generated by them are various and heterogeneous,then when the information of the same item is uploaded to the Internet,there will be a variety of expressions in the end terminal.In order to solve the problem that the machine does not understand the information,Semantic Web of Things(SWoT)emerged with the introduction of semantic technology in IoT.In view of the above-mentioned incorrect expression of resource semantic,this paper uses ontology and link open data to express semantic information in the context of SWoT,and proposes a ontology annotation and named entity combined sensor semantic enhancement method OANESSEM.OANESSEM uses SSN ontology to annotate sensor data to generate machine understandable semantic sensor data,and uses named entity correlation method to semantically enhance sensor data with entities in the linked open data.In the process of semantic enhancement,this paper first extract named entity from semantic sensor data,and then construct named entity candidate sets by using knowledge base DBpedia.Then this paper uses graph-based named entity disambiguation method to calculate the optimal solution of the target entity.The result of semantic enhancement is to obtain the most accurate target entity URI for the named entity from DBpedia,which can be used to describe sensors and enable users to better understand sensor data.Finally,an ontology annotation and named entity combined sensor semantic enhancement system is developed.Firstly,the overall structure of the system is designed,the first-level system data flow diagram and the data flow diagram of each module are given.Then,the system operation process is given and the experimental results are compared with other existing methods.The experimental results show that the semantic enhancement system constructed in this paper can describe the semantics of the sensor well and achieve the purpose of semantic enhancement with improved precision rate and recall rate,thus can be better used to achieve sensor semantic enhancement function.