Method and device for anomaly detection of time series data
By combining representation networks, memory networks, and graph convolutional neural networks, the problem of low accuracy in anomaly detection of time-series data in existing technologies is solved, and efficient anomaly detection of complex time-series data is achieved.
Patent Information
- Application Number
- CN202510067249.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-16
- Publication Date
- 2026-07-17
AI Technical Summary
Existing anomaly detection methods can detect data anomalies at a single point in time, but their accuracy is low when dealing with complex time-series data, especially in scenarios such as automatic fault detection, fault recovery, and system management.
A method based on representation networks and memory networks is adopted to extract features and calculate similarity matrices for time series data of different lengths. By combining graph convolutional neural networks for local convolution calculations, a loss function is constructed and jointly trained to form a time series data anomaly detection model.
It improves the accuracy and robustness of anomaly detection, enhances the ability to process time-series data of different lengths, and strengthens detection performance.
Smart Images
Figure CN122412899A_ABST