Asset change-based multi-dimensional attack surface identification method and system, and storage medium

Abstract

This application relates to the technical field of data analysis, and discloses a method, system, and storage medium for multi-dimensional attack surface identification based on asset changes. The method includes: receiving standardized change events and parsing key attributes from the standardized change events; obtaining business context tags associated with multi-dimensional data assets from an asset knowledge graph; matching control items from a compliance rule set according to the change dimension and change type, and loading the weight values ​​and judgment logic scripts of the control items; obtaining intelligence text associated with the change content snapshot and converting the intelligence text into an intelligence vector; constructing a fusion vector based on the change content snapshot and the intelligence vector; obtaining a compliance status quantification algorithm, constructing a weighted mask vector, and performing a Hadamard product operation on the fusion vector and the weighted mask vector to generate a weighted fusion vector; constructing a structured prompt word template, calling a large language model processing instruction, processing the fusion vector, and outputting the attack surface judgment result, thereby improving the accuracy of risk identification.

Images