A method of natural language understanding on SPARQL ontology query

In this paper, the development of a SPARQL ontology query based on Natural Language Understanding is presented. To obtain ontology knowledge conveniently, Stanford Parser for user´s natural language inquiring is utilized, and query triple according to the grammar is constructed. The method greatly reduces the number of combinations compared with the key word method. Combined with user dictionary, the terms of query triple can be more accurately mapped to the ontology entities. Meanwhile scores calculation is not only considered the similarity of words´ form and semantic, but also considered the ambiguity of concept, for returning to the specific concept as far as possible. Experimental results are presented to demonstrate the performance and validity of method.