An important node mining method and device based on an investment relation network

Abstract

The invention discloses an important node mining method and device based on an investment relationship network, and the device is used for realizing the method, and the method comprises the steps: building the investment relationship network based on an investment behavior; using the robustness index as a measure node importance ranking quality R, and defining the quality R; reversing the currentnetwork maximum connection scale value of the deleted node in the quality R as the current network maximum connection scale value of the added node; calculating the current network maximum connectionscale value of the cost [i] added network access node of the node i by adopting a reverse generation network method BGN, carrying out minimum sorting on the quality R, and mining the first k nodes ofthe sorting sequence as an important node set. According to the method, important nodes are mined by adopting a reverse generation network algorithm, so that the method is more novel, high in precision and high in efficiency.

Description

technical field [0001] The invention relates to the field of data mining, in particular to a method and device for mining important nodes based on investment relationship networks. Background technique [0002] In recent years, as the role of venture capital mechanism in the development of the global economy has become increasingly prominent, the issue of venture capital has become the focus of the capital market, and related research has also emerged. Due to the agency problem and information asymmetry problem, the listed companies in our country will more or less have over-investment or under-investment. It is of great significance to improve the investment efficiency of enterprises. [0003] The investment network is composed of nodes and connections on the network. Under the joint action of nodes and connections, a multi-level complex network is formed. Each node can be used as an individual, group, or organization, and represents stakeholders in the relationship netwo...

AI Technical Summary

Problems solved by technology

Although these two methods consider global information, the time complexity is extremely high, which limits their application in large-scale networks.

PageRank is the core algorithm of the Google search engine. The selection of the parameter c in the algorithm often needs to be obtained through experiments, and the optimal parameters are not universal in different application backgrounds.

Images