Research Of Spatial RDF Data Keyword Search Based On Summary Graph

With the rapid development of Semantic Web and the popularity of mobile terminals,spatial semantic data(spatial RDF data)has increased greatly.Retrieval of spatial RDF data has become a hot research topic.At present,some query languages such as SPARQL and SeRQL support RDF data query,but they are too complex for ordinary users because they require users to master the grammatical rules of query language and schema information of data to be queried.Therefore,RDF spatial keyword query has attracted wide attention from academia and industry.In this context,this paper proposes a keyword query method for spatial RDF data based on summary graph.In the pre-processing stage,several RDF sub-graph structures are extracted according to entity types,and a set of summary graphs is generated.In the query stage,there are two levels of search operations.The first level is the summary graph level search.The reverse search method is used to generate sub-graphs containing all the key words to narrow the scope of the query.Then,the output of the first level search is used as the input of the second level search,and the spatial keyword Top-k query is carried out at the data level.The second level of search is to construct candidate results by BFS from spatial nodes.Finally,the candidate results are sorted and Top-k optimal solutions are output.In the search process,pruning optimization is also carried out to improve query efficiency.In order to evaluate the performance of the algorithm,several control experiments are designed.All experimental data are from real data.Experiments were carried out on experimental data sets of different scales.The algorithm was tested comprehensively and accurately from the aspects of running time and number of access indexes.The experimental results show that the proposed method has better query efficiency.