Method and device for mining frequent sub-graphs of single graph

Abstract

The invention provides a frequent subgraph mining method and device for a single graph, and the method comprises the steps: generating a standard adjacency matrix according to a dictionary sorting result of node labels of the single graph, and numbering the nodes of each graph in sequence; an initial suboptimal normative adjacency matrix tree is generated through the normative adjacency matrix, leaf nodes comprise a first number of edges, and a CSP search space is a dictionary sorting sequence combination of numbers of graph nodes corresponding to node labels contained in the leaf nodes; performing FFSM-Join operation or FFSM-Extension operation on the leaf nodes according to the standard adjacency matrix, and increasing a sub-graph to obtain child nodes of which one edge is expanded; taking the child nodes as candidate sub-graphs, and constructing a CSP search space of the child nodes according to a sub-graph growth mode; if the valid number of the search spaces is smaller than a set support degree threshold value, marking the candidate sub-graphs as invalid sub-graphs; and if the increase is not completed, continuing to perform sub-graph increase, and if the sub-graph increase is completed, outputting a frequent sub-graph. Through the scheme, the frequent subgraph mining efficiency can be improved.

Description

technical field [0001] The invention relates to the technical field of data mining, in particular to a method and device for mining frequent subgraphs of a single graph. Background technique [0002] With the rapid development of big data technology, the use of graph structure to describe data is gradually applied to massive data. Traditional big data analysis technologies usually have a relatively general analysis engine based on SQL or SQL-like tabular analysis tools, but due to the complexity and particularity of relational storage, massive graph data often requires a dedicated calculation and analysis engine to achieve . [0003] A graph is a high-level abstraction of a structure. Frequent subgraph mining is one of the key technologies of graph mining. It has a wide range of applications in social networks, intelligence mining, bioengineering, communication network optimization, text mining, and knowledge reasoning, such as protein structure analysis, link prediction, ...

AI Technical Summary

Problems solved by technology

[0006] However, the existing massive data frequent subgraph mining algorithm engine is not a dedicated engine for graph computing optimization, which greatly affects the mining efficiency of frequent subgraph algorithms

Images