Security Events Graph Alert Prioritization via Chain Aggregation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Security analysts face challenges in identifying and prioritizing critical security alerts within vast amounts of log data generated by enterprise networks, as large graphs representing network entities and alerts become overwhelming, leading to missed important alerts and false positives, especially during cyberattacks that impact multiple entities.

Innovation Solution

The system employs graph traversal and scoring techniques to automatically cluster and prioritize security alerts by propagating native alert scores through the graph, using equivalence and chain collapsing methods to simplify graph structures without hiding high-interest nodes, thereby reducing clutter and focusing analysts on high-risk areas.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If security analysts review all security logs and graphs manually, then they can identify security alerts, but the graphs become overwhelming and analysts miss important alerts

Engineering Contradiction:
Improvealert identification accuracyVSAvoidgraph complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and highlights only the most critical nodes and edges from the complete security graph based on alert scores and propagation analysis. By separating high-priority elements from the rest of the graph, the system presents a simplified view that focuses analyst attention on the most relevant security threats without overwhelming them with the full complexity of the network graph.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies different visual qualities and levels of detail to different parts of the graph based on their security relevance. High-scoring nodes and edges receive enhanced visualization and prioritization, while lower-priority elements are simplified or aggregated. This local differentiation allows analysts to quickly identify critical threats without being distracted by less important graph elements.

Inventive Principle:
Principle #3Local quality

2Loss of information

If the system displays all nodes and edges in the security graph, then complete information is provided, but analysts are overwhelmed and cannot focus on critical alerts

Engineering Contradiction:
Improveinformation completenessVSAvoidanalyst workload
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The patent segments the security graph into multiple priority levels or clusters based on alert scores and propagation analysis. By dividing the graph into hierarchical groups (e.g., high-priority clusters, medium-priority clusters, low-priority clusters), the system maintains complete information in the background while presenting segmented, manageable portions to analysts in the foreground interface.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic graph visualization that adapts to analyst interactions and threat severity. The graph can dynamically expand or collapse clusters, adjust visualization density, and reprioritize elements based on real-time alert propagation and analyst focus areas. This dynamic behavior allows the system to maintain information completeness while adapting the presentation complexity to current analytical needs.

Inventive Principle:
Principle #15Dynamics

3Measurement precision

If graph traversal propagates scores through all nodes, then comprehensive alert prioritization is achieved, but processing time increases

Engineering Contradiction:
Improvealert scoring accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary scoring and filtering operations on security events before full graph propagation. By pre-calculating basic alert scores and identifying high-priority starting nodes in advance, the system reduces the scope and complexity of the subsequent graph traversal and score propagation process, achieving comprehensive analysis with reduced processing time.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS20240137390A1Security Events Graph for Alert Prioritization
Publication Date: 2024.04.25 NETSKOPE INC
  • US20240137390A1 patent drawing
  • US20240137390A1 patent drawing
  • US20240137390A1 patent drawing

AI summary

The technology disclosed includes a system to reduce clutter when displaying a security analysis graph of nodes and edges. Simple chains of nodes do not have branches and are equivalent when they have the same length, connection types and endpoints. First, second and potentially more simple chains can be aggregated for display. A third and potentially more simple chains can be excluded from aggregation based on an accumulated risk analysis score. The excluded simple chain can readily be called to an analyst's attention.