Power law attribute network geometry generation method

The invention provides a power law attribute network geometry generation method. The method comprises the steps of fitting attribute distribution of a data set; fitting network node degree distribution, and determining distribution parameters; candidate nodes in the hyperbolic space are randomly generated, and hyperbolic distances between the candidate nodes are calculated; and calculating a connection probability between the candidate nodes, and generating a connection edge according to the connection probability between the candidate nodes. According to the method, the geometric characteristic of the generation model is guaranteed, namely, the nodes are located in the high-dimensional hyperbolic space, and based on the characteristic, the generation model well utilizes the node space structures such as distance, cluster relation and position distribution to analyze the cause of network evolution to form a visual global network overview, so that the network evolution efficiency is improved. And an analysis m