Graph search algorithm applied to large-scale irregular structure data

Abstract

The invention provides a graph search algorithm applied to large-scale irregular structure data. The algorithm includes a data preprocessing method and an inquiry executing method. The data preprocessing method includes the steps of unifying the formats of the irregular structure data, establishing a neighbor label vector table for original image points of all images, and establishing an expansion image with property points. The inquiry executing method includes the steps of screening out candidate matching points corresponding to key points in original image data points through a neighbor label, calculating the matching degree of the candidate matching points and selecting the central point of a local area, dividing the local area on the periphery of the central point, and inquiring about sub-images to be matched with similar images of local images. By means of the algorithm, search accuracy is ensured, calculation complexity is greatly decreased, and feasible and effective image search of large-scale irregular structure data can be achieved.

Description

technical field [0001] The invention belongs to the fields of graph data management and graph data mining, and relates to a graph search algorithm applied to large-scale irregular structure data. Background technique [0002] The graph model is one of the most common structural models, which is produced by natural or artificial structures and is widely used in the description of various relationships. Among them, points are usually used to represent entities, such as molecules, individuals, electronic accounts, etc.; edges are usually used to describe the specific relationship between two entities, such as molecular bonds, communication, etc. [0003] In social computing, bioinformatics and other applications, there is a large amount of irregular structure data processing requirements, and mining algorithms based on graph models have gradually attracted people's attention. Among them, a kind of non-regular structure data that uses discretized labels to identify attributes a...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problems that the existing graph search algorithm has high complexity and efficiency cannot meet the requirements, the present invention proposes a graph search algorithm applied to large-scale irregular structure data, which reduces the complexity of the algorithm under the condition of ensuring accuracy degree, improving the efficiency of graph search

Images