Alert Generation Model Using Historical Login Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cloud-based IT monitoring systems overwhelm administrators with notifications, leading to missed important alerts and unnecessary alerts, due to manual and experience-based condition selection, which is burdensome and error-prone, ignoring collective administrator experience.
Innovation Solution
Implementing a data analytics-based system that receives device state data and administrator login data to apply a model, determining when to send alerts based on historical data, automatically identifying when a device requires service, reflecting the collective experience of all administrators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual rules are used to determine alert conditions, then the system is simple to implement, but the accuracy of alert prioritization is poor and important alerts may be missed
Solution Approach 1:
The system automatically generates alert rules by analyzing historical administrator behavior data without requiring manual configuration. The model self-learns from patterns in the data, automatically determining which device states typically require administrator intervention, thereby eliminating manual rule creation while maintaining high accuracy
Solution Approach 2:
The patent replaces manual mechanical rule-creation processes with an automated data analytics system. Instead of administrators manually creating and maintaining rules, the system uses computational algorithms to analyze historical data and generate optimized alert rules automatically
2Measurement precision
If manual experience-based rules are used for alert conditions, then implementation is straightforward, but collective administrator experience is ignored leading to suboptimal alert selection
Solution Approach 1:
The system merges the individual experiences of multiple administrators into a single collective intelligence model. By analyzing login data from many administrators, the system captures diverse expertise and synthesizes it into optimized alert rules that reflect collective wisdom rather than individual experience
Solution Approach 2:
The system automatically acquires and processes historical login data to build the predictive model without requiring manual input of expert knowledge. The model self-configures by learning from patterns in the data, making the system easy to operate while achieving high accuracy through collective experience
3Reliability
If alerts are sent for all possible device issues, then comprehensive monitoring is achieved, but administrators receive overwhelming numbers of alerts causing them to ignore important ones
Solution Approach 1:
The system extracts only the most relevant alert conditions by analyzing which device states historically led to administrator logins. It filters out unnecessary alerts and focuses only on states that typically require intervention, thereby reducing alert volume while maintaining detection of important issues
Solution Approach 2:
The system uses historical login data as feedback to continuously refine alert rules. By analyzing patterns of when administrators actually accessed devices, the model learns which alerts are most valuable and adjusts its predictions to maximize the relevance of sent alerts
Data Source
AI summary
Improved techniques of identifying when a device needs of service involve using data analytics to determine conditions when a device administrator of a computerized device is to be sent an alert regarding that device. Along these lines, a device monitoring system receives state data from a device that indicates the device is in a particular state, e.g., running certain applications, using some percentage of the processor and memory capacity, etc. The device monitoring system maps the device state data to a decision of whether or not to send a device administrator an alert concerning the device. The decision itself is a result of applying a model to the state data that is derived from the application of data analytics on historical device state data and administrator login data.