Training Data Label Verification Using Unsupervised Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Supervised machine learning models are unreliable due to labeling errors in training data, leading to inaccurate outputs and operational inefficiencies.
Innovation Solution
A computing platform applies unsupervised learning algorithms to identify discrepancies between clustering and labeling information, flags data labeling errors, and trains supervised learning models using corrected data, while grading data sources based on accuracy to improve model reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If supervised machine learning models are trained using labelled data, then the models can make automated predictions, but labeling errors in the training data reduce reliability and result in inaccurate outputs
Solution Approach 1:
The patent introduces unsupervised learning algorithms as an intermediary mechanism to verify labeling accuracy. The system applies unsupervised clustering to training data, compares clustering results with provided labels, and identifies discrepancies. This intermediary verification process detects labeling errors without requiring additional labeled data, thereby improving model reliability while maintaining the use of labeled training data.
2Measurement precision
If data points with labeling errors are removed from the training set, then model accuracy improves, but the quantity of training data decreases
Solution Approach 1:
The system performs self-verification by applying unsupervised learning algorithms to its own training data to identify labeling errors. The unsupervised clustering process automatically compares data groupings with provided labels, enabling the system to self-diagnose and self-correct training data quality issues without external verification, thus maintaining high accuracy while preserving maximum training data volume.
3Reliability
If unsupervised learning algorithms are applied to verify labeling accuracy, then data quality improves, but computational complexity and processing time increase
Solution Approach 1:
The system applies unsupervised learning algorithms selectively rather than to all data uniformly. It focuses the computational effort on verifying labeling accuracy for training data points where discrepancies are most likely to impact model performance, performing partial verification that achieves sufficient data quality improvement without the full computational burden of complete verification.
Data Source
AI summary
Aspects of the disclosure relate to improving training data used for model generation. The computing platform may receive, from one or more data sources, a labelled data set. The computing platform may apply, to the labelled data set, an unsupervised learning algorithm, which may result in a clustered data set corresponding to the labelled data set. The computing platform may compare, for each data point in the labelled data set, corresponding clustering information and labelling information to identify discrepancies. The computing platform may flag, for data points with identified discrepancies between the corresponding clustering information and labelling information, a data labelling error. Using data points without identified discrepancies between the corresponding clustering information and labelling information, the computing platform may train a supervised learning model. The computing platform then may store the trained supervised learning model.