Event Graph Narrative Generation for Network Root Cause Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network event management tools struggle to understand complex event graphs, interpret implicit information, and provide adequate responses to network events, often failing to address underlying causes and considering relationships across multiple domain topologies.
Innovation Solution
A computer program product utilizing a large language model with topological context adapters processes event graphs and associated text to generate narratives and remediations, incorporating training data from past situations and actions, to provide actionable insights and solutions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If complex event graphs are used to represent causal chains of events, then the completeness of event representation is improved, but the interpretability and understanding of events deteriorates
Solution Approach 1:
The patent introduces a large language model as an intermediary between the complex event graph and human users. The LLM processes the graph data and generates natural language narratives that explain the causal chains of events in human-understandable terms, thereby maintaining complete event representation while improving interpretability
Solution Approach 2:
The patent replaces manual analysis of event graphs with automated natural language generation using large language models. This substitution transforms the mechanical process of graph interpretation into an intelligent system that automatically produces explanatory narratives, resolving the contradiction between completeness and interpretability
2Device complexity
If existing network event management tools are used, then the simplicity of the system is maintained, but the ability to understand and respond to network events deteriorates
Solution Approach 1:
The patent enhances existing event management tools by integrating a large language model that provides multiple functions: understanding event graphs, generating narratives, identifying root causes, and suggesting remediations. This multi-functional approach improves event understanding capability while building upon existing simple tool frameworks
3Measurement precision
If manual analysis of event graphs is performed, then the accuracy of event interpretation is improved, but the productivity of event response deteriorates
Solution Approach 1:
The patent implements an automated system where the large language model performs event graph analysis independently without requiring manual human intervention. The system self-generates narratives, identifies root causes, and proposes remediations, thereby maintaining high interpretation accuracy while dramatically improving response productivity through automation
Data Source
AI summary
Described systems and techniques determine an event graph of a causal chain of events representing a situation within a network, the event graph including event text characterizing at least one event of the causal chain of events. The event graph may then be processed using a large language model that includes at least one topological context adapter that includes a graph adapter and a text adapter, including processing the event graph with the graph adapter and the event text with the text adapter. The at least one topological context adapter may be trained using existing narratives describing past situations, and/or may be trained using worklog data describing past situations and corresponding actions taken to remedy the past situations. Outputs of the graph adapter and the text adapter may be combined to generate a narrative of the situation that explains the causal chain of events and/or instructions to remedy the situation.