Managing and Mining Graph Data part 52 is a comprehensive survey book in graph data analytics. It contains extensive surveys on important graph topics such as graph languages, indexing, clustering, data generation, pattern mining, classification, keyword search, pattern matching, and privacy. It also studies a number of domain-specific scenarios such as stream mining, web graphs, social networks, chemical and biological data. The chapters are written by leading researchers, and provide a broad perspective of the area. This is the first comprehensive survey book in the emerging topic of graph data processing. . | Graph Mining Applications to Social Network Analysis 499 Clearly c nasi-elic ue becomes a clique when 7 1. Note tti t this densitybased group typically dots nos guarantee the nodal degree or readability for each node in the group. St allows the degree of different nodes to vary drasti-cdly ihus seamt more suitable for large-scale networks. In l the maximum .-dense quasi-cliquos are tiasplooed. A greedy algorithm io adopted to idnd - maximal quasi-clique. The quasi-clique is initialized wtth it vcrlrx with the largest cleg ice tn the network. and then expanded with nodet that are 1 ikety lo contriSiuic to a iauge qua-i-clique. This expansion continúen untiO no nodes can loe added to maintain the 7-densityi lividcntly. this itiisccit sorrel for maximal quasi-clique its not optimal. So a subsequent local search procedure fGRAtSP is applied So find a larger maximal quasi-clique in the local neighboxhootl This procedure it tittle do detect a close-to-optimal maximal quasitdique but requires the whois graph to be in main memory. To handie -aogefScale nctworkr. ilie authoos proposed to utilize the procedure above Io find out 1Sos Sower Sound of degreet lor pruning. In each iteration a cuCscI of edges are sampled from tha networks and GRASP is applied to find a locally maximal quasiidlque. Si ipicosc the quasi-clique is of size k it is im-poscihle to include in mssximai quari-cliquc a node with degree less than ky all of whose ceii littors also have their degree less than ky. So the node aod its infiden- edges cao te pruned from the graph. This pruning process is repealed until GRASP caa ae applieO directly So the remaining graph to find out time maximaS quasi-clique. Pot a directed graili like 1tac Web. tise work tn 19 extends the complete-btpartite core 29 to 7-siente bipartite. X Y is a 7-dense bipartite if .x G X X x n Y y YI Vy G Y X- y n X 7 X where 7 ami 7 are user psovidod constants. OXs authors derive a heuristic to islficti ntioi prune the .
