The Research And Application Of Based On Ontology Multi-source Heterogeneous Data Fusion Method

At present,some progress has been made in the field of data fusion between multi-source and heterogeneous data whether in academia or industry.However,there are still many problems on the semantic fusion of multi-source and heterogeneous data.For example,in the field of multi-sensor data fusion,the fusion obstacle brought by the fusion system due to the characteristics of multi-source,heterogeneity,and incompleteness of sensor data.How to make use of the constraint relationship between these multi-source data and transform the multi-source heterogeneous data into a consistent interpretation and semantic description of the sensor's monitoring target at the semantic level.On the basis of this,a more accurate and reliable fusion scenario description of the situation is obtained.Finally,according to the specific situation assessment results automatically given the auxiliary decision-making information,these issues are the focus of this study.In view of the above problems,this paper presents a method of ontology-based multi-source heterogeneous data fusion based on the existing research and validates the proposed method for the case of multi-sensor data fusion in the field of the smart home.The main research contents of this paper are as follows:Firstly,aiming at the problem of heterogeneity in the process of multi-sensor data fusion,this paper constructs a multi-sensor ontology description model.Through the unified ontology description of multi-sensor data,the problem of semantic heterogeneity and grammatical heterogeneity of multisource heterogeneous data is solved,and the efficiency of data fusion is improved.In order to fusion the instance data of the sensor ontology in feature level,an update algorithm of the sensor ontology instance and an algorithm of attribute fusion and feature extraction are proposed.Secondly,aiming at the problem of multi-sensor data fusion in the field of smart home,a data fusion model in the field of the smart home is proposed.At the frame level,the technical integration of data fusion in the field of the smart home is carried out,and then the situation ontology in the field of the smart home is constructed according to the proposed framework.In order to correlate feature-level ontology with situation ontology,this paper adopts ontology mapping method.Aiming at the uncertainty characteristics of multi-sourced data,a decision-level fusion method is proposed,which uses the transferable belief model to fusion the multi-frame information and finally generates the situation assessment results under the current smart home environment.On the basis of this,some data that is more sensitive to timeliness are processed in a time-sensitive manner.After obtaining the result of the situation assessment,the method of ontology reasoning is used to generate the decision information automatically.Finally,in order to verify the validity of the proposed method and implement decision-level data fusion system in the smart home environment,a DFS prototype tool for data fusion system in the field of the smart home is designed and implemented.The DFS prototype tool is used to accomplish the mapping between feature-level ontology and situation ontology,and the consistency of ontology is tested.Finally,rule-based ontology reasoning is carried out based on the results of the situation assessment to automatically generate decision information.