Merge clustering-based parallel frequent probability subgraph searching method

Abstract

The invention discloses a merge clustering-based parallel frequent probability subgraph searching method. In allusion to the problem that the existing frequent subgraph searching method is high in time-space consumption and cannot satisfy the big data environment requirement, the method comprises the following steps: mapping a probability subgraph into circuit topology to be processed by adopting a node voltage method; clustering the probability subgraph by utilizing merge clustering so as to effectively reduce the time expenditure; and finally realizing the method on the basis of a Spark frame, so as to further improve the calculation speed and expandability.

Description

technical field [0001] The invention relates to a parallelized frequent probability subgraph mining method based on merging and clustering, which is used to realize rapid mining and identification of frequent probability subgraphs in a large-scale probability network, and belongs to the technical field of computer data mining. Background technique [0002] With the continuous emergence of information and Internet technologies, a large amount of network data has been generated, such as social networks, biological networks, etc., and these networks can be represented by graph models. How to efficiently implement data mining on graph datasets has become one of the hot issues in the field of data mining research. In practical applications, many graph data exist in the form of probability. For example, in the field of bioinformatics, the biological network data obtained by research usually have inevitable experimental errors or noise data; at the same time, the process of biolog...

AI Technical Summary

Problems solved by technology

[0005] Purpose of the invention: Aiming at the problems and deficiencies in the prior art, the present invention provides a parallelized frequent probability subgraph search method based on merge clustering, which can effectively solve the problem of using simple hierarchical clustering while ensuring calculation accuracy. The class handles the problem of excessive time and space complexity of large-scale probability graph data

Images