Automatic construction method of information security knowledge graph based on deep learning

Abstract

An automatic construction method of an information security knowledge graph based on deep learning comprises the following steps: step 1, constructing an ontology database of the information securityknowledge graph; step 2, labeling the collected data based on a heuristic rule mode to obtain a training set of entity recognition, designing a model structure based on Bi-LSTM for learning to serve as a model of entity recognition, and extracting entity words of the text; and step 3, performing automatic labeling based on triple knowledge to obtain a training set of relationship extraction, and designing an attention mechanism-based LSTM structure as a model of relationship extraction so as to extract triples (relationships) in the text.

Description

technical field [0001] The invention relates to an automatic construction method of a map, in particular to an automatic construction method of an information security knowledge map based on deep learning. Background technique [0002] Most of the current information security knowledge bases are maintained and updated manually, such as the well-known CVE (Common Vulnerability Knowledge Base), but most of the updates of CVE are manually updated, and some newly discovered sites cannot be updated in time. Vulnerabilities and attack knowledge, and these knowledge are often published in the blogs of relevant companies, manufacturers or security teams at the first time, so it is necessary to adopt an automatic update method to mine and extract knowledge from text. [0003] In addition to the public knowledge base, there are some existing studies such as Stucco, which use machine learning methods to extract triples from text to construct knowledge graphs, and use maximum entropy mo...

AI Technical Summary

Problems solved by technology

[0004] However, there are two problems in the current existing technology. First, although the method based on manual sorting has a good accuracy rate, there is a problem that the update is not timely; second, the current research based on machine learning methods, the effect of the algorithm Not good enough, especially in the relationship extraction stage, learning is only possible with human assistance, and the construction process is still not fully automated

Images