A research on linked data-based Chinese semantic retrieval model

Abstract

The overall objective of this study is to implement the semantic retrieval model, and provide the linked information of retrieval result. The work presented in this paper focuses on the core technique, the method used in our study is to translate the Chinese natural language to SPARQL, and the procedure we followed can be briefly described as the several aspects. First, we get the dependency relationship via language technology platform, and then we find the 'head' (or topic) of the sentence. Third, we analyze the relationship between modified words and modifiers of the question with the knowledge base. Finally, we build the RDF triples and compose the SPARQL query, and discover the linked data of retrieval result. We carried out the example in this paper to experiment to test the validity of the method. The studies we have performed showed that the relevent of the model is better than the traditional information retrieval; what's more, this work also provided the linked information, which supported users to learn more knowledge conveniently. © 2014 Springer-Verlag Berlin Heidelberg.