Scalable association rule mining with predication on semantic representations of data

Finding semantic associations from a vast amount of heterogeneous data is an important and useful task in various applications. We present a framework to extract semantic association patterns directly from a very large graph dataset without the extra step of converting graph data into transaction data. The proposed algorithm SAG (Semantic Association Generator) utilizes the principle of minimum description length to unearth general relevant associations and demonstrates that SPARQL commands can be used to perform data mining tasks.