Supporting the efficient exploration of large-scale social networks for recommendation

ALEJANDRO CORBELLINI; CRISTIAN MATEOS; DANIELA GODOY; ALEJANDRO ZUNINO; SILVIA SCHIAFFINO

Maceió

Workshop; Proceedings of the II Brazilian Workshop on Social Network Analysis and Mining (BraSNAM 2013); 2013

SBC

Most recommendation algorithms in the context of large-scale social networks struggle with the need of an efficient exploration of the underlying user graph. Current solutions in the form of graph-specific databases or frameworks for graph algorithms do not scale well for processing complex navigational patterns. In this paper we present an approach for supporting social recommendation algorithms that operate with large graphs in a computer cluster based on "policies", rules that allow users to throttle the amount of parallelism and control task location.