clique算法的基本思路

clique算法的基本思路内容摘要：

its, the candidate kdimensional units are determined using candidate generation procedure.  MDLbased pruning  To decide which subspaces(and the corresponding dense units) are interesting.  MDLMinimal Description Length candidate generation procedure  Input: Dk1, the set of all (k1)dimensional dense unit  Output: a superset of the set of all kdimensional dense units  Algorithm: MDLbased pruning  Coverage of subspace sj  Sort the subspaces in the descending order of their coverage  Divide the sorted list of subspaces into two sets: the selected set I and the pruned set P  How to arrive at the cut point MDLbased pruning  The code length is minimized to determine the optimal cut point i MDLbased pruning 第二步：识别聚类  Input: a set of dense units D, all in the same kdimensional space S  Output:a partition of D into D 1,…,D q,such that all units in D i are connected and no two units u iD i, u jD j with ij are connected. Each such partition is a cluster  Method: depthfirst search algorithm  Start with some unit u in D, assign it the first cluster number,and find all the units it is connected。