Secondary ML Model Revising Primary Predictions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning models face challenges in accurately classifying unlabeled structured data, such as distinguishing between columns of phone numbers and bank account numbers, leading to inefficiencies and reduced data utilization due to the lack of metadata and low detected probabilities.
Innovation Solution
A secondary machine learning model is used to revise predictions from a primary model by implicitly learning metadata and utilizing the structure of the input data, such as patterns in a data array, to increase the probability of correct classifications and decrease incorrect ones, thereby improving data type detection and security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single machine learning model is used for data classification, then the system complexity is low, but the classification accuracy is insufficient
Solution Approach 1:
The patent divides the classification system into two separate models: a first machine learning model that performs initial classification, and a second machine learning model that performs revision classification. This segmentation allows each model to specialize in specific tasks, improving overall accuracy while managing complexity through functional division.
Solution Approach 2:
The first machine learning model acts as an intermediary between the raw data and the second machine learning model. It processes the unlabelled structured data and provides preliminary classifications that the second model then refines, serving as a bridge that improves the final classification accuracy.
2Productivity
If machine learning models process unlabelled structured data without metadata, then data utilization is limited, but adding metadata increases data processing complexity
Solution Approach 1:
The machine learning models perform self-service by automatically learning metadata patterns from the structure of unlabelled data. The models identify relationships between data entries and infer classification rules without requiring external metadata, enabling the system to utilize data effectively while avoiding the complexity of manual metadata management.
3Measurement precision
If the first machine learning model makes initial predictions, then processing speed is fast, but prediction accuracy is low
Solution Approach 1:
The first machine learning model performs preliminary classification actions on unlabelled data, providing initial predictions that are then refined by the second model. This preliminary action allows the system to quickly process data initially, while the second model subsequently improves accuracy for critical classifications without requiring all data to undergo extensive processing.
Solution Approach 2:
The second machine learning model applies revision actions selectively rather than to all data points. It focuses computational resources on cases where the first model's predictions are uncertain or likely to benefit from revision, performing partial action that improves accuracy while minimizing additional processing time.
Data Source
AI summary
Various embodiments are generally directed to techniques for prediction based machine learning (ML) models, such as to utilize a ML model to generate predictions based on the output of another ML model. Some embodiments are particularly directed to a secondary ML model that revises predictions generated by a primary ML model based on structured input data. In many embodiments, the secondary ML model may utilize predictions from the primary ML model to learn metadata regarding the structured input data. In many such embodiments, the metadata regarding the structured input data may be used to revise the predictions from the primary ML model. For example, the secondary ML model may utilize a structure of the input data combined with patterns in the predictions from the primary ML model to revise the predictions from the primary ML model.