On mining cross-graph quasi-cliques

摘要: Joint mining of multiple data sets can often discover interesting, novel, and reliable patterns which cannot be obtained solely from any single source. For example, in cross-market customer segmentation, a group customers who behave similarly markets should considered as more coherent cluster than clusters found market. As another bioinformatics, by joint gene expression protein interaction data, we find genes show also produce interacting proteins. Such may potential pathways.In this paper, investigate novel problem, cross-graph quasi-cliques, is generalized several interesting applications such segmentation data. We build general model for why the complete set quasi-cliques previous methods, study complexity problem. While problem difficult, develop an efficient algorithm, Crochet, exploits effective techniques heuristics to efficaciously mine quasi-cliques. A systematic performance reported on both synthetic real sets. demonstrate some meaningful bioinformatics. The experimental results that algorithm Crochet scalable.