Method and device for mining sub-graphs in uncertain data interaction graph

Abstract

The invention discloses a method and a device for mining sub-graphs in an uncertain data interaction graph, the method comprises the following steps: inputting an uncertain graph, degree limitation and a probability critical value, the uncertain graph being used for representing the relationship among a plurality of nodes with unstable connection relationship, and the degree limitation and the probability critical value being used for obtaining the sub-graphs satisfying specific limitation; pruning is carried out on the uncertain graph; solving the occurrence probability of each node of the pruned uncertain sub-graph; and traversing each node in the pruned uncertain sub-graph, and removing the node from the pruned uncertain sub-graph when the occurrence probability of the node is smaller than a probability critical value. By adopting the method, the fluctuation of the sub-graph probability caused by the instability of data in a real scene can be fully considered, and the sub-graph mining stability is improved; and meanwhile, three pruning algorithms are used, so that the calculation amount is reduced, and the resource consumption is greatly reduced compared with other sub-graph mining algorithms in the same mining scene.

Description

Technical field [0001] The present invention relates to the field of data mining, in particular to a method and apparatus for mining (k, θ) - clique in an uncertain graph. Background [0002] In recent years, global big data has entered a period of accelerated development, the amount of data has shown exponential and explosive growth, and the data generated by the interaction between different individuals in these large amounts of data is expressed in the form of graphs. Therefore, mining subgraphs from a complex network becomes an important topic. At the same time, the cluster is the most widely used sub-graph model, which is used to find overlapping communities in social networks, detect social classes, and identify protein complexes in protein interactions. Therefore, how to efficiently excavate sub-diagrams from the map has become a major challenge that all sectors of society need to solve. [0003] From the past few years, existing research has focused mainly on the enumera...

AI Technical Summary

Problems solved by technology

[0005] However, the existing subgraph mining technology has the problems of poor stability of the subgraph and large amount of calculation in the mining process.

Images