Asset Risk Assessment Using Attack Depth for Automated Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional asset risk assessment methods rely on expert knowledge for determining feature weights, leading to inefficiencies and inaccuracies in security evaluations due to the subjective nature of these judgments.
Innovation Solution
An asset risk assessment method that utilizes an attack depth analysis of alert messages to objectively determine security feature values and weights, enabling automated risk assessments without relying on manual experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If expert knowledge is used to determine feature weights, then the assessment can be performed, but the efficiency and accuracy are reduced due to subjective judgment and manual processes
Solution Approach 1:
The patent replaces the manual expert judgment process with an automated machine learning system. The model automatically learns feature weights from historical data and attack depth information, eliminating the need for human experts to manually determine weights. This substitution of mechanical/manual process with automated intelligent system directly addresses the contradiction by improving both efficiency (automated processing) and accuracy (data-driven weight determination).
Solution Approach 2:
The patent changes the parameter determination approach from static expert-defined weights to dynamic data-driven weights. The feature weights are no longer fixed by expert knowledge but are automatically adjusted based on attack depth levels and historical assessment data. This parameter change enables the system to adapt to different attack scenarios while maintaining high accuracy and efficiency.
2Productivity
If manual expert judgment is used for feature weight determination, then the assessment process can be implemented, but the process becomes slow and lacks focus
Solution Approach 1:
The system performs self-service by automatically determining feature weights without requiring external expert intervention. The machine learning model autonomously learns from historical data and attack depth information to establish appropriate weights for each feature. This self-service capability dramatically increases assessment speed while the automated nature of the process actually reduces operational complexity despite the sophisticated underlying algorithms.
3Reliability
If feature weights are determined by expert knowledge, then the assessment can be performed, but the results are directly affected by the magnitude of feature weights leading to low accuracy
Solution Approach 1:
The patent implements feedback mechanisms where the machine learning model continuously learns from historical assessment data and attack outcomes. The system uses feedback from actual attack results and assessment outcomes to iteratively improve feature weight determination. This feedback loop ensures that the assessment results are not arbitrarily affected by weight magnitudes but are instead grounded in actual security performance data, thereby improving both reliability and precision.
Data Source
AI summary
An asset risk assessment method, an apparatus, a computer device, and a storage medium are provided. The asset risk assessment method includes: receiving an alert message of an asset to be assessed; obtaining an attack depth of the alert message based on the alert message, and the attack depth being a degree to which the asset to be assessed is subjected to attack; obtaining a security feature value of the asset to be assessed; obtaining a feature weight of the security feature value based on the attack depth; and performing a risk assessment of the asset to be assessed based on the security feature value and the feature weight thereof, to obtain a risk assessment result.