SocialGQ: Towards semantically approximated and user-Aware querying of social-graph data

Abstract

The proliferation of social and collaborative sites makes users increasingly active in the generation of socialgraph data; however, such sea of data often hinders them from finding the information they need. In this paper, we present SocialGQ ("Social-Graph Querying"), a novel approach for the effective and efficient querying of socialgraph data overcoming the limitations of typical search approaches proposed in the literature. SocialGQ allows users to compose complex queries in a simple way, and is able to retrieve useful knowledge (top-k answers) by jointly exploiting: (a) the structure of the graph, semantically approximating the user's requests with meaningful answers; (b) the unstructured textual resources of the graph; (c) its social and user-Aware dimension. An experimental evaluation comparing SocialGQ to leading approaches shows strong gains on a real social-graph data scenario.