Research On Mining Algorithm Of Tight Sub-graphs In Uncertain Graphs

With the rapid development of social science and technology,a large number of data in graphical form have been accumulated in various industries.Such as,in bio-protein networks,social networks,and wireless sensor networks,a large amount of data is generated,which contains a lot of useful knowledge and potentially valuable information,so the mining technology based on graph structure emerges as the times require.Graph mining refers to the process of discovering valuable information from the data of these graph nodes.So far,domestic and foreign scholars have carried out extensive research on the problem of graph node data mining,but it is difficult to pass the commonly used due to the connection uncertainty between graph nodes,the sparsity of graphs and the number of graph nodes.The data structure is analyzed,so in order to satisfy the problem of analyzing data in the above network,a discovery algorithm is implemented.Firstly,for the fast mining algorithm of compact subgraph in dense sub-graphs of uncertain graphs,the function of the ratio of the number of edges to the number of vertices in the subgraph and the expected degree of each vertex in the subgraph are used to measure the tightness.Implementation,each time you need to delete a node and connected edges,resulting in algorithm time redundancy.Therefore,in order to reduce the algorithm time overhead,according to the relationship of the decision node connection and the calculation of the vertex degree,the ImproveALKS algorithm with the minimum vertex degree is not unique.Secondly,in order to obtain the problem of multiple subgraph regions with close correlation and high probability of occurrence in the large image,and the input of the threshold parameters of the compact subgraph is adjusted according to the connection condition of the vertices in different graphs,and then the GC-ImproveALKS algorithm is obtained.The Top-k non-overlapping compact subgraph is obtained through greedy thought,and the compact subgraph is found based on the algorithm optimization strategy.Then delete all the nodes and edges of the subgraph,and then calculate the compact subgraph of the remaining nodes,and the cycle calculation is carried until the k value is satisfied.Finally,the time efficiency of the compact subgraph mining algorithm and the Top-k subgraph mining algorithm is verified by comparison experiments of multiple sets of data.