ML Anomaly Detection for Time-Based Metric Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to efficiently detect anomaly conditions in time-based metric data from computer devices, leading to potential harm to software, hardware, or users, and requiring lengthy timeframes for root cause identification and resolution.
Innovation Solution
A computer system and method utilizing machine learning techniques, specifically an anomaly detection AI/ML model, to identify early warning indicators of anomalies in time-based metric data. This involves training ML models for each device to determine threshold operating values, comparing real-time data to these thresholds, and notifying support teams to initiate remedial actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional anomaly detection methods are used, then system administrators can monitor network traffic, but the time required to identify and resolve anomalies is lengthy
Solution Approach 1:
The system performs preliminary actions by training machine learning models on historical data before anomalies occur. The models learn normal behavior patterns in advance, enabling them to quickly identify deviations when anomalies happen, thus reducing detection time while maintaining accuracy.
Solution Approach 2:
The patent replaces manual mechanical analysis methods with machine learning-based automated detection. The ML models automatically analyze time-based metric data, substituting the manual monitoring and analysis process, which significantly reduces the time required to identify and resolve anomalies while improving detection accuracy.
2Measurement precision
If machine learning models are trained for each device, then anomaly detection accuracy improves, but system complexity increases
Solution Approach 1:
The system segments the anomaly detection problem by training individual machine learning models for each device. This segmentation allows each model to specialize in detecting anomalies specific to its device's behavior patterns, improving detection accuracy while managing complexity through modular, device-specific models rather than one complex system.
Solution Approach 2:
Each device has its own trained ML model that autonomously monitors and detects anomalies in its time-based metric data. The models self-service by automatically learning from historical data and independently identifying anomalies without requiring centralized complex coordination, thus improving accuracy while keeping system complexity manageable.
Data Source
AI summary
A computer implemented method and system for detecting one or more anomaly conditions in one or more computer devices. A Machine Learning (ML) model is trained for each of the one or more computer devices to determine threshold operating values for time-based metric data associated with each of the one or more computer devices. Utilizing the trained ML model, time-based metric data is compared for each of the one or more computer devices to the determined threshold operating values to determine if the time-based metric data falls outside of the determined threshold operating values. Provided is notification of an anomaly condition for a computer device responsive to determining the time-based metric data falls outside of the determined threshold operating values associated with the computer device.