Anomaly Detection for Time Series Data Without Fixed Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly detection methods for time series data lack flexibility and accuracy, requiring fixed threshold values and label data, which are costly and have low scalability.
Innovation Solution
An anomaly detection method that obtains feature information from time series data, generates a target feature combination, and performs anomaly detection without requiring fixed threshold values or label data, using an electronic device with a processor and memory to execute the method.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fixed threshold values and label data are used for anomaly detection, then detection accuracy can be improved, but the cost increases and scalability decreases
Solution Approach 1:
The system performs self-service by automatically generating feature combinations and detecting anomalies without requiring external label data or manual threshold setting. The anomaly detection model learns from the feature combinations and autonomously identifies anomalies, eliminating the need for costly labeled datasets and manual calibration while maintaining high accuracy and scalability
Solution Approach 2:
The system changes parameters by dynamically generating feature combinations based on time series data characteristics rather than using fixed threshold values. The feature combination generation process adapts to different data patterns, allowing the system to maintain high detection accuracy across diverse scenarios without requiring re-labeling or manual threshold adjustment for each new dataset
2Ease of operation
If fixed threshold values are used for anomaly detection, then the detection process becomes simpler, but flexibility and accuracy are reduced
Solution Approach 1:
The system performs preliminary action by pre-generating multiple feature combinations that capture different aspects of time series data patterns. These feature combinations are prepared in advance through automated feature engineering, allowing the detection process to directly compare against learned patterns without requiring manual threshold configuration, thus maintaining both simplicity and accuracy
Solution Approach 2:
The feature combination generation mechanism serves multiple functions: it automatically creates diverse feature representations, adapts to different data types and patterns, and provides a unified framework for anomaly detection across various scenarios. This universal approach replaces the need for separate manual threshold settings for different cases, maintaining operational simplicity while improving accuracy
Data Source
AI summary
An anomaly detection method for time series data includes: obtaining a plurality of pieces of feature information of time series data; generating a target feature combination based on the plurality of pieces of feature information; and performing anomaly result detection based on the target feature combination.