Automated Incident Configuration Linking via Feature Vector Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Incident reports in remote network management platforms often go unlinked to configuration items due to human error or undiscovered devices, making it difficult for agents to assist users effectively.
Innovation Solution
A method involving a computational model that generates feature vectors from incident reports, compares them to existing vectors, and automatically links relevant configuration items, ensuring accurate and efficient linking of configuration items to incident reports.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual linking of configuration items to incident reports is performed, then agents can accurately identify relevant devices, but the process is time-consuming and error-prone
Solution Approach 1:
The system automatically performs the linking of configuration items to incident reports using machine learning models and feature vector comparison, eliminating the need for manual agent intervention. The computational instance self-services by discovering devices, generating feature vectors from incident reports, comparing them against known devices, and automatically establishing links without human input.
Solution Approach 2:
The manual mechanical process of agents searching and linking configuration items is replaced with an automated computational system using machine learning models, feature vector generation, and algorithmic comparison. This substitutes human cognitive and manual operations with automated information processing and pattern recognition.
2Productivity
If automated methods are used to link configuration items to incident reports, then time is reduced, but accuracy may deteriorate due to human error or undiscovered devices
Solution Approach 1:
The system performs preliminary device discovery and configuration item registration before incident reports are generated. By pre-establishing a comprehensive database of known devices with their feature vectors, the system ensures that when an incident report arrives, the matching process can immediately compare against an up-to-date inventory, reducing the chance of linking errors or missing undiscovered devices.
Solution Approach 2:
The system incorporates feedback mechanisms where agents can review and correct automated linking suggestions, and where the system learns from correction patterns. This feedback loop continuously improves the accuracy of the machine learning models and feature vector comparisons, ensuring that productivity gains do not compromise reliability.
3Measurement precision
If comprehensive device discovery is performed to ensure all configuration items are known, then linking accuracy improves, but system complexity and resources increase
Solution Approach 1:
The system performs device discovery to a sufficient degree rather than attempting exhaustive discovery. By using feature vector comparison and pattern matching, the system can accurately link incident reports to known devices without needing to discover every possible device in the network. This partial action approach achieves adequate completeness while avoiding the exponential complexity of exhaustive discovery methods.
Data Source
AI summary
Visualizations are automatically generated based at least on a received identifier that identifies a report. One or more data sets and one or more style configurations are retrieved from a shared visualization library via a backend API based on the received identifier and passed to a router component. The router component includes a switch that renders a score component, which generates a score visualization, if the retrieved data sets include score data. If the data sets include time series data, or the score visualization also includes a time series graph, the router component renders a time series component that passes the data sets and the style configurations to a parser that parses the data sets and the style configurations and outputs one or more highcharts options objects. A highcharts component generates the time series visualizations and/or adds the time series graph to the score visualization based on the highcharts options objects.