Dynamic link prediction device based on generative adversarial network and application

Abstract

The invention discloses a dynamic link prediction device based on a generative adversarial network. The device comprises a computer memory, a computer processor and a computer program. A dynamic linkprediction model is stored in the computer memory. The computer processor implements the following steps when executing the computer program: inputting a dynamic network to be predicted into a dynamiclink prediction model, and outputting a dynamic network structure at the next moment through calculation, the dynamic network comprising a power grid, a computer network, a social network, a trafficnetwork, a biological network, an aviation network and a reading community network. The dynamic link prediction device can generate the prediction network close to the real situation, and the effect of dynamic link prediction is achieved. The invention also discloses an application of the dynamic link prediction device in an aviation network, a social network and a reading community network.

Description

technical field [0001] The invention belongs to the security field, and in particular relates to a dynamic link prediction device and application based on a generative confrontation network. Background technique [0002] Many complex systems in reality can be modeled as networks, such as power grids, computer networks, social networks, transportation networks, biological networks, etc. Through the analysis and prediction of the network structure and state, it can provide reference for the control of traffic congestion, the development of new targeted drugs, and the research and development of a new generation of cellular networks. As a tool for network analysis, link prediction aims to predict unknown potential link states based on known nodes and links. Most existing link prediction methods mainly focus on static networks, extracting and learning network structures without temporal information. [0003] Traditional link prediction methods are static link prediction method...

AI Technical Summary

Problems solved by technology

Traditional static network link prediction methods cannot effectively solve this kind of link prediction problems in real networks with time variables.

Images