Anomaly Detection Model Retraining From Worker Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for detecting anomalous states in plants using unsupervised machine learning struggle to improve precision due to unlabeled pre-collected data, making it difficult to evaluate the learning model and determine its appropriateness.
Innovation Solution
A detection apparatus and method that collects measurement data, inputs it into a trained learning model, and generates a retrained model by using user-provided label values and collected measurement data when the detection result differs from the user's determination, thereby improving the model's precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If unsupervised machine learning is performed on pre-collected normal data to generate a normal state model, then the model can be created without labeled data, but the precision of the learning model to detect abnormality predictors is difficult to improve
Solution Approach 1:
The system implements feedback by collecting determination results from users who check measurement devices and comparing them with detection results from the learning model. When discrepancies are found, the system automatically retrain the model using the user's determination results as labels, creating a continuous improvement loop that enhances detection precision over time
Solution Approach 2:
The system performs preliminary actions by automatically collecting measurement data and preparing training datasets before retraining is needed. The system proactively identifies when retraining should occur by comparing detection results with user determinations, and pre-prepares the training data using collected measurement data and user feedback
2Measurement precision
If labeled data are used to evaluate and improve the learning model, then precision can be improved, but the complexity of the system increases due to manual labeling requirements
Solution Approach 1:
The system performs self-service by automatically collecting measurement data, preparing training datasets, and retraining the learning model without requiring manual intervention for data collection or model retraining. The system autonomously identifies when improvement is needed and executes the retraining process using collected data and user feedback as labels
Solution Approach 2:
The system uses feedback from user determinations to automatically improve the model. User feedback on detection accuracy triggers automatic retraining processes, creating a self-improving system that reduces the need for manual model development and evaluation iterations
Data Source
AI summary
A server device collects measurement data measured by a field device, inputs the measurement data collected, into a trained detection model that predicts a predetermined event in response to input of measurement data, obtains an output result from the trained detection model, executes, in a case where the output result from the trained detection model is different from a determination result by a worker W who has checked the field device for which the predetermined event was predicted, retraining of the trained detection model by using a label value input by the worker W and the measurement data collected, and thereby generates a retrained model.