| Popis: |
The problem of graph summarization has practical applications involving visualization and graphcompression. As graph-structured databases become popular and large, summarizing andcompressing graph-structured databases can become more and more useful. We explore the use of aparticular family of graph summarization algorithms we call Summaries with Supernodes,Superedges, and Corrections (SSSC) and the feasibility of using SSSC algorithms whensummarizing large Resource Description Framework (RDF) graph datasets. We also proposeoptimizations to the Uniform Randomized SSSC algorithm by using trained heuristics to pick seednodes. We also show how SSSC summaries may be stored in a similar manner as RDF triple stores,and we discuss possibilities for future work involving localized SSSC algorithms. |
