A Hybrid Linked Data Query Execution Approach using Backlinks

Typical Linked Data Query Execution (LDQE) approaches do not take into consideration the availability of backlinks, resulting in an incomplete set of query results. In this paper, we provide a detailed discussion of existing techniques for query processing over LD and propose a framework for LDQE based on backlinks, which can yield missing interlinking information for a given LD SPARQL query. The proposed approach is hybrid in nature, an index is maintained for storing information of backlinks which are extracted from a simulated LD cloud. These backlinks are then classified into a number of classes. Additionally, live exploration is performed using link traversal technique for retrieving more URIs over the Web of LD.