Research On Subgraph Mining Method For Social Networks

With the development of internet technology,more and more people are using online social platforms for socializing,which has rapidly expanded the scale of social networks.Community subgraph partitioning for large-scale social networks has become one of the mainstream research directions.The subgraph mining algorithm is one of the important means for community subgraph partitioning.The existing subgraph mining algorithms have the disadvantage of rarely considering the degree of relationship between nodes in social networks,and the low efficiency of subgraph structure mining calculations for large-scale social networks makes it an important research direction to improve the effectiveness and computational efficiency of subgraph mining algorithms.Therefore,this article proposes a subgraph mining algorithm for social networks.The main work of the subgraph mining algorithm for social networks is: first,the relationship degree between nodes in the social network graph is abstracted into weights and weights are defined,and the first round of social network graph is divided based on this;Second,the second round of division is carried out for the subgraph after the first round of division,and the ownership of the subgraph nodes is adjusted by using the modular optimization verification,and the result subgraph set is obtained.Thirdly,aiming at the problem of low computational efficiency of subgraph mining of large-scale social network datasets,a dynamic resource scheduling algorithm for social networks is proposed,which improves the computational efficiency of subgraph mining by improving the cluster resource scheduling strategy.In the dynamic resource scheduling algorithm,the real-time available resources of the compute nodes in the cluster are first defined,and then the resources required for subgraph computing are combined with the available resources of each node in the cluster,and the subgraph is allocated to the most suitable cluster computing nodes for calculation.Finally,experimental results on datasets of different scales demonstrate that the subgraph mining algorithm proposed in thesis for social networks can achieve better results compared to the comparison algorithm,and can effectively improve computational efficiency on largescale social network datasets.