Two-step RDF query processing for Linked Data

Since RDF triples are modeled as graphs, we cannot directly adopt existing solutions from relational databases and XML technologies. Thus, there are still a number of open problems in the area of Linked Data. We present a hybrid method between centralized and distributed approaches. By using auxiliary indexes based on the MBB approximation, our approach can retrieve distributed Linked Data efficiently. The goal of our approach is to support efficient join query processing by quickly pruning unnecessary scanning data.