3D Anomaly Dashboard for Network Behavior Pattern Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional user behavior analysis (UBA) struggles to identify critical behavior anomalies across large enterprises due to the difficulty in analyzing every account's behavior, especially with hundreds to thousands of users, and generalized organizational rules fail to detect deviations in typical yet potentially malicious patterns.

Innovation Solution

A digital fingerprinting (DFP) workflow using artificial intelligence and machine learning to generate fine-grained, unsupervised behavior models for each account, identifying user-specific anti-patterns through a suite of techniques and pipelines that analyze day-to-day activities, enabling early alerting on anomalous behavior.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional user behavior analysis uses generalized organizational rules to identify patterns, then it can detect explicitly bad patterns, but it fails to detect deviations in typical yet potentially malicious patterns and cannot deeply analyze every account behavior across large enterprises

Engineering Contradiction:
Improveanomaly detection precisionVSAvoidanalysis system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the analysis system into multiple specialized components: unsupervised behavior models for each account, supervised anomaly detection models, and a prioritization system. This segmentation allows each component to focus on specific tasks, improving overall detection precision while managing complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimension of analysis by moving from rule-based detection to machine learning-based behavioral analysis. It adds temporal dimensions through time-series analysis and multi-dimensional scoring (anomaly score, risk rating, prioritization score) to detect subtle deviations that traditional rules miss.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If the system analyzes every account login or application activity across an enterprise network with hundreds to thousands of users, then it can identify critical behavior anomalies, but it becomes difficult to rapidly identify and react to threats

Engineering Contradiction:
Improvethreat detection reliabilityVSAvoidthreat response speed
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary actions by pre-training unsupervised behavior models for each account using historical data. These models establish baseline behaviors in advance, enabling rapid real-time anomaly detection without requiring complex analysis during threat events. The prioritization system also pre-ranks anomalies to speed up response.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent replaces manual security analysis with automated machine learning systems. Supervised anomaly detection models and automated prioritization algorithms substitute for manual inspection, enabling rapid processing of large volumes of account activities while maintaining high detection reliability.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If traditional UBA systems detect anomalies across large enterprises, then they can identify security threats, but they generate excessive false positives that increase the load on cybersecurity analysts

Engineering Contradiction:
Improveanomaly detection accuracyVSAvoidsignal-to-noise ratio
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The system implements feedback loops where supervised anomaly detection models are trained on confirmed anomalies, and the prioritization system learns from analyst actions. This continuous feedback improves detection accuracy and reduces false positives over time by refining what constitutes a true anomaly versus normal variation.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent changes detection parameters by using multiple scoring dimensions (anomaly score, risk rating, prioritization score) instead of single threshold rules. It dynamically adjusts sensitivity based on account-specific baselines and contextual factors, reducing false positives while maintaining detection of true anomalies.

Inventive Principle:
Principle #35Parameter changes

4Loss of information

If the system provides detailed analysis of every account behavior, then it can deeply understand user patterns, but it becomes difficult to rapidly identify critical behavior anomalies for security analysts

Engineering Contradiction:
Improvebehavioral pattern informationVSAvoidanalyst workflow efficiency
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The system extracts only the most critical information from comprehensive account behavior analysis. The prioritization system identifies and extracts top anomalies requiring analyst attention, while less critical patterns are handled automatically or deferred. This extraction maintains detailed behavioral understanding while presenting only essential information to analysts.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent adds a prioritization dimension to the analysis output, transforming comprehensive behavioral data into ranked anomaly lists. It uses multi-dimensional scoring (anomaly score, risk rating, business impact) to project detailed analysis results into a format that highlights critical issues without overwhelming analysts with raw data volume.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP4339814B1Visualization technology for finding anomalous patterns
Publication Date: 2025.12.10 NVIDIA CORP
  • EP4339814B1 patent drawingFigure 1
  • EP4339814B1 patent drawingFigure 2
  • EP4339814B1 patent drawingFigure 3

AI summary

Technologies for generating a graphical user interface (GUI) dashboard with a three-dimensional (3D) grid of unit cells are described. An anomaly statistic can be determined for a set of records. A subset of network address identifiers can be identified and sorted according to the anomaly statistic. The subset can have higher anomaly statistics than other network address identifiers. There can be a maximum number in the subset. The GUI dashboard is generated with unit cells organized by the subset of network address identifiers as rows, time intervals as columns, colors as a configurable anomaly score indicator, and a number of network access events as column heights. Each unit cell is a colored, 3D visual object representing a composite score of anomaly scores associated with zero or more network access events corresponding to the respective network address identifier at the respective time interval. The GUI dashboard is rendered on a display.