Context-Aware Anomaly Detection Using Recurrent Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly detection methods require manual user input for context recognition and anomaly identification, making them inefficient and dependent on user classification.
Innovation Solution
A computer-implemented method using recurrent neural networks and dynamic time warping to automatically detect anomalies by generating prediction models from training data and comparing them with real-time data, reducing the need for user input through context-aware wavelet analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual user input is used for context recognition and anomaly identification, then user control and interpretation accuracy are maintained, but efficiency is reduced and user dependence increases
Solution Approach 1:
The system performs self-service by automatically recognizing contexts and identifying anomalies without requiring manual user input. The recurrent neural network autonomously processes time-series data to detect anomalies, eliminating the need for users to manually classify or identify anomalies, thereby resolving the contradiction between automation extent and detection efficiency
Solution Approach 2:
The patent replaces the mechanical manual process of context recognition and anomaly identification with an automated computational system using recurrent neural networks. This substitution transforms the manual mechanical operation into an automated intelligent system, improving efficiency while reducing user dependence
2Reliability
If traditional anomaly detection methods are used, then implementation simplicity is maintained, but false positives and negatives increase
Solution Approach 1:
The patent segments the time-series data into contextual units that can be independently analyzed. By dividing the continuous data stream into manageable segments with defined contexts, the system can apply targeted anomaly detection algorithms to each segment, improving accuracy while managing complexity through structured data organization
Solution Approach 2:
The system performs preliminary context recognition and data preprocessing before anomaly detection. By pre-processing the time-series data to establish contexts and patterns in advance, the system reduces the complexity of the actual anomaly detection process while improving reliability through prepared contextual frameworks
Data Source
AI summary
A computer-implemented method for context aware anomaly detection includes receiving a signal including real-time data, dividing the received signal into a set of training wavelets bounded by a respective training context, generating a prediction model for the training context by processing the set of training wavelets through a recurrent neural network to identify patterns in a sequence of data of a respective training wavelet. The method includes, in response to execution of a current context that is different from the training context, extracting current wavelets from a signal of the current context, identifying a pairwise match between same dimensions of the current wavelets and a wavelet pattern of the prediction model by aligning sequences of the current wavelet and sequences of the wavelet pattern of the prediction model. Upon determining that a total cumulative distance between the two sequences exceeds a predetermined threshold, an anomaly is identified.