K-Core algorithm is a subgraph structure, which is used to find closely related subgraphs that meet the specified kernel degree. In the resulting subgraph of k-kernel, each vertex has at least k degrees, and all vertices are connected with at least k other nodes in the subgraph. K-Core is usually used to divide a graph into subgraphs. By removing unimportant vertices, outdated subgraphs are exposed for further analysis. Because of its linear time complexity and intuitive interpretability, k-Core has many application scenarios in risk control finance, social networking, biology and other fields.

The process of K accounting method is also very simple. A * * * is divided into two steps. In fact, the contents of the two steps are the same. As for why the same process should be executed in two steps, you can think for yourself.

In Figure 2, we give a simple 3-nucleon graph partition process.

K-accounting method is usually used to find subgraphs that meet the specified k-kernel degree in the graph, and subgraphs occupy the core position in the graph. The higher the kernel degree, the smaller the subgraph and the greater the corresponding kernel degree. In a sense, the subgraph divided by the kernel degree plays an important role in the original graph, such as tracing the origin and evolution trend of the graph and identifying the intermediary of the graph. There are many specific application scenarios. You can refer to a paper: k-core: Theory and Application.