Network security entity overlapping relation identification method, device and equipment

Abstract

The embodiment of the invention discloses a network security entity overlapping relation identification method, device and equipment, and the method comprises the steps: obtaining to-be-identified text data, and inputting the to-be-identified text data into a target joint extraction model; the network security entity vocabularies in the to-be-recognized text data output by the target joint extraction model and the entity relationship between the network security entity vocabularies can be obtained, and then the entity relationship between the target network security entity included in the to-be-recognized text data and the network security entity vocabularies is determined. According to the target joint extraction model, parameter and structure sharing in the network security entity identification and relation extraction process can be realized, and the correlation between network security entity identification and entity relation extraction is enhanced. The transmission of errors of entity identification is reduced, the accuracy of network security entity identification and entity relationship extraction is improved, and the extraction of the overlapping relationship between the network security entities is realized.

Description

technical field [0001] The present application relates to the technical field of network security, in particular to a method, device and equipment for identifying overlapping relationships of network security entities. Background technique [0002] In the existing network environment, there are many factors that threaten network security. Accurately discovering network security entities and their relationships from network information can effectively analyze the status quo of current security issues and the development trend of future security issues, and make corresponding security decisions accordingly. [0003] In network texts, there are some words and phrases related to network security threats, and such words and phrases are called network security entities. By performing network security entity recognition on related texts, the network security entities contained in the text can be obtained. In addition, specific relationships may exist between network security enti...

AI Technical Summary

Problems solved by technology

[0004] At present, the accuracy of network security entity identification and entity relationship extraction in network information text is not high, and the "one-to-many" overlapping relationship between network security entities in network information text cannot be completely extracted. As a result, network security threat information cannot be obtained timely and accurately from network information, which restricts network security situation analysis and early warning capabilities

Images