Graph Convolutional Anomaly Detection via Feedback Retraining
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly detection systems using graph convolutional neural networks face inefficiencies and unreliability for real-time or near-real-time predictive tasks due to the structural complexity of graph-based databases, hindering the effective utilization of graph-based data for anomaly detection.
Innovation Solution
Retraining the graph convolutional neural network model using confirmation feedback data to enhance real-time accuracy and dependability without requiring expensive redesign or offline training, by generating confirmation feedback data objects and integrating them into the anomaly detection process, and determining ground-truth anomaly designations for training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If graph convolutional neural network models are used for anomaly detection, then detection capability is improved, but system complexity and training cost increase
Solution Approach 1:
The system performs self-updating by automatically integrating confirmation feedback data into the training set and retraining the model without requiring manual intervention or offline redesign. The anomaly detection system serves itself by continuously improving its own accuracy through automated feedback integration.
Solution Approach 2:
The system implements a feedback loop where anomaly confirmations are collected, integrated into the training data, and used to retrain the model. This closed-loop feedback mechanism allows the system to continuously improve detection accuracy based on actual confirmation outcomes.
2Measurement precision
If offline training is performed to improve model accuracy, then detection precision is improved, but operational time and cost increase
Solution Approach 1:
The model training process is made continuous rather than periodic or offline. The system continuously integrates new confirmation feedback data and retrains the model in real-time, ensuring the detection precision is maintained and improved without stopping operations or incurring offline training delays.
3Reliability
If confirmation feedback data is integrated in real-time, then model accuracy is improved, but processing overhead increases
Solution Approach 1:
The system automatically manages the feedback integration process without requiring external intervention. It self-updates by seamlessly incorporating confirmation data into the training set and triggering retraining workflows autonomously, minimizing the impact on processing efficiency while maintaining high detection reliability.
Data Source
AI summary
There is a need for more effective and efficient anomaly detection. This need can be addressed by, for example, solutions for performing/executing graph convolutional anomaly detection. In one example, a method includes identifying related graph database input data associated with a predictive entity; generating related graph feature data for the predictive entity; generating, based on the related graph feature data and using a graph convolutional neural network model, an anomaly detection score for the predictive entity, wherein at least a portion of the graph convolutional neural network model is trained using confirmation feedback data; performing an anomaly confirmation to generate the confirmation feedback data object for the predictive entity, and integrating the confirmation feedback data object for the predictive entity into the confirmation feedback data associated with the graph convolutional anomaly detection.