Objectives In the era of big data, geospatial information resources are constantly increasing. However, the current knowledge bases are difficult to take into account the semantic knowledge as well as to retrieve geospatial data quickly. Therefore, it is urgent to introduce ontology technology to improve the access speed of geospatial data on the basis of semantic knowledge so as to accurately obtain the information needed by users.
Methods We propose a fast retrieval method that takes into account the semantic knowledge of geospatial data based on ontology. Firstly, a semantic knowledge base is constructed through encoding rules of generic place name, geospatial data and Baidu encyclopedia open source data. Secondly, we design some query rewrite rules and mapping methods between semantic knowledge and spatial database. So that we can construct retrieval statements based on place names automatically.
Results The feasibility of the fast retrieval method is verified through the application of an example and the comparative analysis of efficiency and quality.
Conclusions Based on the semantic information of geospatial data, we can automatically construct the structured query language(SQL) retrieval statement only by geographical entity name, reduce the user's dependence on data storage method and SQL grammar rules, and effectively improve the retrieval efficiency and intelligence of the system.