Target object recognition method and device, computer equipment and storage medium

Abstract

The invention relates to a target object recognition method and device, computer equipment and a storage medium, and relates to the technical field of network security. The method comprises the following steps: initializing a PageRank value of each node in a network relation graph; iteratively calculating according to a random walk algorithm to obtain a similarity value between each node in the network relation graph and the seed node set; according to the similarity value between each node and the seed node set, identifying a target node with the same node type as at least one seed node fromeach node; and determining the network object corresponding to the target node as the target object. According to the scheme, the target nodes of the same type as the seed nodes can be screened out through iterative computation only by specifying a small number of seed nodes of a specific type in advance in the network relation graph, the network objects corresponding to the target nodes are recognized as the target objects, and the accuracy of network object recognition is improved.

Description

technical field [0001] The present application relates to the technical field of network security, in particular to a target object identification method, device, computer equipment and storage medium. Background technique [0002] With the continuous development of network technology, network security technology is getting more and more attention. Among them, how to identify possible malicious users (or devices used by malicious users) in the network is an important research direction in network security technology. [0003] In related art, machine learning models are usually trained to identify malicious users or devices in a network. For example, developers pre-collect characteristic information of marked malicious users, such as user attributes of malicious users, network community attributes, etc., and then train machine learning models based on the characteristic information of malicious users; The feature information is input into the trained machine learning model,...

AI Technical Summary

Problems solved by technology

[0004] However, the recognition accuracy of machine learning models in related technologies depends on the scale of training data, and the number of malicious users or devices in the network is rare, so it is difficult to collect a sufficient amount of training data, resulting in low accuracy of trained machine learning models. Low, poor identification of malicious users or devices

Images