Parallel graph abstracting method based on attribute graph

Abstract

The invention belongs to the technical field of computer graph abstraction, and particularly relates to a parallel graph abstraction method based on an attribute graph, which comprises the following steps of: 1, preprocessing acquired graph data, and processing each node in a graph into a node structure with own information and all direct neighbor information; 2, randomly selecting a direct neighbor node for the current node, and then selecting a node with the same attribute and maximum similarity as the current node from all the direct neighbor nodes of the neighbor node as a candidate node combined with the current node; 3, judging whether an introduced error exceeds an error threshold value or not after the current node and the candidate nodes are combined, if yes, returning to the step2 to continue to search for other candidate nodes, and if not, combining the two nodes; and 4, executing two-node combination by updating all node information in the node structure, repeating the steps 3-4 until the number of the remaining nodes is smaller than a set threshold, storing a final node structure, and exporting an abstract graph.

Description

technical field [0001] The invention belongs to the technical field of computer graph summarization, in particular to a property graph-based parallel graph summarization method. Background technique [0002] Graphs have powerful intrinsic advantages and are widely used in modeling real-world objects and their relationships. Large-scale graph data is common in many application domains. In a graph, entities are modeled as vertices, while their relationships or connections are represented by edges. Various modern applications generate a large amount of graph data. Since a large amount of relational information is encoded in the graph, potential implicit knowledge can be mined from these graph data, so as to better serve users. Therefore, many researchers Both have done in-depth research on the processing and calculation of graph data. However, as the number of application users continues to grow and the scale and structure of graphs become increasingly complex, it becomes a ...

AI Technical Summary

Problems solved by technology

The greedy method is to select the optimal 2-hop neighbor node pair in the whole graph for merging each time, although it selects the best node pair for merging and obtains the smallest summary error, but it causes a lot of calculation and network communication

The random method is to randomly select 2-hop neighbor node pairs each time as candidate node pairs. Although the amount of calculation in the node pair selection stage is greatly reduced, the selected node pairs do not meet the error threshold for node merging with a high probability, resulting in subsequent stage unnecessary calculations

Images