Random Indexing for Searching Large RDF Graphs

Querying large RDF spaces with traditional query languages such as SPARQL is challenging as it requires a familiarity with the structure of the RDF graph and the names (URIs) of its classes, properties and relevant individuals. In this paper, we propose a complementary approach based on Vector Space Models (VSM), more concretely Random Indexing (RI) [1] for building a semantic index for a large RDF graph. Traditionally, a semantic index captures the similarity of terms based on their contextual distribution in a large document collection, and the similarity between documents based on the similarities of the terms contained in them. By creating a semantic index for an RDF graph, we are able to determine contextual similarities between graph nodes (e.g., URIs and literals) and based on these, between arbitrary subgraphs. These similarities can be used for finding a ranked list of similar URIs/literals for given input term which can be used for exploring the repository or enriching SPARQL queries.