RAQ: Relationship-Aware Graph Querying in Large Networks

摘要: The phenomenal growth of graph data from a wide variety real-world applications has rendered querying to be problem paramount importance. Traditional techniques use structural as well node similarities find matches given query in (large) target graph. However, almost all existing have tacitly ignored the presence relationships graphs, which are usually encoded through interactions between and edge labels. In this paper, we propose RAQ-Relationship-Aware Graph Querying-to mitigate gap. Given graph, RAQ identifies k best matching subgraphs that encode similar To assess utility paradigm for knowledge discovery exploration tasks, perform user survey on Internet Movie Database (IMDb), where an overwhelming 86% 170 surveyed users preferred relationship-aware match over traditional querying. need subgraph isomorphism renders NP-hard. is made practical beam stack search. Extensive experiments multiple datasets demonstrate effective, efficient, scalable.