A Natural Language Query Method for Linked Data
Zheng Xiao, Yang Xiao · 2021
Efficient data retrieval is one of the key issues in the development of Web of Data based on Resource Description Framework (RDF). Formal query language like SPARQL is an effective way to retrieve structured data, but users are accustomed to natural language-based retrieval. Due to its grammatical complexity and prerequisite knowledge of ontology schema, the formal query language is hard to be applied in natural language-based retrieval directly without help from new approaches or tools. Therefore, how to automatically convert keyword search into formal query-based retrieval is an important part of the realization of Web of Data. The natural language query method for linked data automatically converts natural language queries into SPARQL queries to improve the effectiveness and efficiency of the system. In this paper, the SPARQL query is generated by transforming language elements among abstract models, followed by constructing the ontology-based query semantic graphs and illuminating semantic ambiguity. The experimental results show that the proposed approach has a higher recall rate, better precision, and lower time consumption.