Causal Event Graph Feedback for Continuous Knowledge Graph Generation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing network event management tools struggle to analyze, respond to, predict, or prevent undesired events due to the difficulty in determining relationships between numerous events across complex and dispersed network topologies, often requiring manual adjustments of clustering parameters that become inaccurate over time.

Innovation Solution

A computer program product generates a knowledge graph using a machine learning model to process causal graph feedback and spatiotemporal context, enabling automated identification of causal relationships between events, root cause analysis, and predictive remediation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If manual adjustments of clustering parameters are used to analyze events, then initial analysis capability is provided, but accuracy deteriorates over time and requires continuous manual intervention

Engineering Contradiction:

Improveautomated identification of causal relationshipsVSAvoidaccuracy of event relationship analysis

Core Design Contradiction:

Extent of automationVSMeasurement precision

Solution Approach 1:

The system implements feedback loops where the knowledge graph continuously learns from new events and feedback signals. The machine learning model processes feedback on causal graphs and updates the knowledge graph accordingly, enabling the system to improve accuracy over time through automated learning rather than manual parameter adjustments.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The knowledge graph system performs self-service by automatically identifying causal relationships between events using machine learning. The system autonomously updates its internal representations and improves its analysis capabilities without requiring continuous manual intervention, thereby maintaining high accuracy while achieving full automation.

Inventive Principle:

Principle #25Self-service

2Measurement precision

If all possible relationships between numerous events are examined, then comprehensive analysis is achieved, but processing time becomes infeasible

Engineering Contradiction:

Improvecomprehensiveness of event relationship analysisVSAvoidprocessing time for event analysis

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system segments the complex task of analyzing all possible event relationships into manageable components by organizing events into a structured knowledge graph. The machine learning model processes events incrementally and identifies causal relationships selectively rather than examining all possible combinations, reducing processing time while maintaining comprehensiveness.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary action by pre-processing events and organizing them into the knowledge graph structure before full analysis. The machine learning model prepares causal graphs and spacetime context in advance, enabling faster subsequent analysis of event relationships without requiring exhaustive examination of all possibilities in real-time.

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If complex clustering parameters are used to capture event relationships, then analysis accuracy improves, but system complexity increases

Engineering Contradiction:

Improveaccuracy of event clusteringVSAvoidcomplexity of clustering parameters

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system replaces complex mechanical clustering parameter adjustments with a machine learning-based knowledge graph system. Instead of manually configuring and tuning clustering parameters, the machine learning model automatically learns event relationships and causal patterns from data, simplifying the system while improving accuracy through adaptive learning.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

4Productivity

If existing network event management tools are used, then basic event detection is provided, but predictive and preventive capabilities are insufficient

Engineering Contradiction:

Improveevent detection capabilityVSAvoidpredictive and preventive capability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The knowledge graph system enables predictive capabilities by performing preliminary action - it continuously learns from historical events and causal relationships to predict future events before they occur. The machine learning model analyzes patterns in the knowledge graph to anticipate potential issues, allowing preventive actions to be taken before problems manifest, thereby improving reliability while maintaining detection productivity.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12505360B2Continuous knowledge graph generation using causal event graph feedback

Publication Date: 2025.12.23 BMC HELIX INC

AI summary

Described systems and techniques determine causal associations between events that occur within an information technology landscape. Individual situations that are likely to represent active occurrences requiring a response may be identified as causal event clusters, without requiring manual tuning to determine cluster boundaries. Consequently, it is possible to identify root causes, analyze effects, predict future events, continuously generate a knowledge graph, and prevent undesired outcomes, even in complicated, dispersed, interconnected systems.