BRDPHHC: A Balance RDF Data Partitioning Algorithm Based on Hybrid Hierarchical Clustering

Data partitioning is a fundamental step to achieve effective storage and query of RDF big data. This paper presents a balance RDF data partitioning algorithm based on hybrid hierarchical clustering (BRDPHHC), which combines AP and K-means clustering. BRDPHHC´s functionality includes three aspects: (i) a pre-processing step combining nodes compression and nodes remove to reduce the scale of raw data points, (ii) AP clustering algorithm is used to coarsen the RDF graph step by step and produce data blocks, and (iii) K-means algorithm is used for data partitioning finally. Experiments on benchmark datasets demonstrate the effectiveness of the proposed scheme.