Ranking the Results of DBpedia Retrieval with SPARQL Query

摘要

In recent years, a number of Semantic Web databases have been actively open to public as common resources on the Web by the effort of Linked Open Data community project. Due to this, we need a good method to search necessary data from those databases, depending on various purposes. In this study, we propose two methods to rank the results retrieved by a SPARQL query (especially, a SELECT query), using the information about the frequency of each property in a data set and the links between RDF resources. In order to evaluate our proposed methods, we set two cases for using SPARQL queries, and then rank the query results in each case. The usefulness of our proposed method has been confirmed by subject experiments.