Network Alert Response Automation via Learned User Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Responding to system alerts in network environments is labor-intensive for users, as existing automation systems require users to codify responses to incoming events, placing a significant burden on them.
Innovation Solution
A method where a collection system learns from user responses to system indicators, creating alert event entries and automatically generating responses to new indicators that match previous alerts, reducing user intervention through a learning module and automatic processing module.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If users manually respond to system alerts, then response accuracy is maintained, but labor intensity and time consumption increase significantly
Solution Approach 1:
The system automatically learns from user responses and generates responses autonomously without requiring users to manually codify rules. The collection system observes user actions, creates alert event entries, and automatically applies learned responses to matching alerts, enabling the system to serve itself rather than requiring continuous user intervention.
Solution Approach 2:
The system implements a feedback loop where user responses to alerts are observed and stored as alert event entries. This feedback is used to continuously improve and refine automatic responses through pattern recognition and matching, allowing the system to learn from past user decisions and apply them to future similar situations.
2Extent of automation
If automation systems are implemented, then user burden is reduced, but users must codify responses which requires significant initial effort
Solution Approach 1:
Instead of requiring users to codify responses upfront and then applying them automatically, the system inverts the approach by first observing user responses and then automatically generating the codification. The system learns patterns from user behavior and creates the automation rules itself, reversing the traditional automation setup process.
Solution Approach 2:
The system performs preliminary learning and pattern recognition by observing user responses to various alerts before full automation is needed. Alert event entries are created in advance through observation, building a knowledge base that enables future automatic responses without requiring users to prepare codification rules beforehand.
3Reliability
If users manually interpret and respond to each alert, then response accuracy is maintained, but time consumption increases
Solution Approach 1:
The system creates copies of successful user responses by storing them as alert event entries with associated parameter types and values. When similar alerts occur, the system retrieves and applies these copied responses, maintaining the accuracy of proven user decisions while eliminating the need for users to re-interpret and re-create responses for similar situations.
Data Source
AI summary
In a method for providing an automatic learned response in a network, a collection system observes user responses to the incoming system indicators and to parameter types and associated parameter values used in the user responses. The collection system creates alert event entries to includes the incoming system indicators, confidence thresholds, the user responses, and the parameter types and associated parameter values used in the user responses. When the collection system receives new system indicators, the collection system determines whether the new system indicators match the system indicators in one or more alert event entries. When the new system indicators match the system indicators in one or more alert event entries and the confidence level exceeds the confidence threshold, the collection system automatically creates a new response based on the user response, the parameter types, and the associated parameter values in the matching alert event entries.