Progression Machine Learning Model for Classification Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing predictive data analysis solutions face challenges in improving training speed without compromising predictive accuracy, leading to inefficiencies in computational operations and data requirements for machine learning models.
Innovation Solution
A method involving a progression machine learning model that predicts the progression of classification by using a base cohort dataset with initial and subsequent severity level labels, allowing for the generation of predictive outputs that enhance predictive accuracy and reduce the need for computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing predictive data analysis solutions are used to improve training speed, then training efficiency increases, but predictive accuracy deteriorates
Solution Approach 1:
The patent segments the training process into two distinct phases: a base cohort training phase using a limited, carefully selected subset of data, and a subsequent fine-tuning phase. This segmentation allows the model to achieve high predictive accuracy on core patterns without requiring extensive computational resources, thereby resolving the contradiction between training speed and accuracy.
Solution Approach 2:
The patent applies preliminary action by pre-selecting a base cohort with specific characteristics that are most representative of the target distribution. This preliminary curation of training data ensures that the model learns from high-quality, relevant examples from the outset, achieving accurate predictions without needing to process large volumes of less relevant data, thus improving training efficiency while maintaining accuracy.
2Measurement precision
If more computational operations and data entries are used, then predictive accuracy improves, but computational efficiency deteriorates
Solution Approach 1:
The patent extracts and utilizes specific, high-value features from the input data that are most predictive of the target outcome. By identifying and focusing on these critical features rather than processing all available data equally, the model achieves high predictive accuracy with reduced computational operations, resolving the contradiction between accuracy and computational complexity.
Solution Approach 2:
The patent applies local quality by assigning different levels of processing and attention to different portions of the data based on their predictive value. The base cohort and selected features receive intensive processing to ensure accuracy, while less critical data elements undergo lighter processing, thereby optimizing the balance between predictive accuracy and computational resource utilization.
3Reliability
If more data entries are stored for training, then model performance improves, but storage efficiency deteriorates
Solution Approach 1:
The patent performs preliminary action by pre-curating a base cohort that contains only the most representative and informative data entries for training. This preliminary selection process identifies and retains high-value data while discarding redundant or less relevant entries, ensuring that the model achieves optimal performance with a minimized dataset, thereby resolving the contradiction between model performance and storage requirements.
Data Source
AI summary
Various embodiments of the present disclosure provide methods, apparatus, systems, computing devices, computing entities, and/or the like for predicting progression of condition classifications using a progression prediction machine learning model. The progression prediction machine learning model is trained using training data that assigns an outcome label to each entity that is in a defined base cohort based at least in part on whether the entity has subsequent severity level that exceeds an initial severity level. Once trained the progression prediction machine learning mode is configured to predict a severity level escalation probability in a future time period for an entity.