Temporal Personal Data Training With Update Logging and Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine learning models for personal data, such as healthcare provider information, suffer from stale training data due to temporal variations, leading to inconsistencies and errors, and lack effective methods to handle fraudulent claims, which are resource-intensive to detect.
Innovation Solution
A system that monitors and logs data updates over time, verifies data accuracy, and uses a scheduler to manage computing resources, ensuring that training data remains current and normalized, employing various machine learning algorithms to predict accurate personal data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual verification of personal data is performed to ensure accuracy, then data reliability is improved, but labor costs and time consumption increase significantly
Solution Approach 1:
The system enables self-service by automatically monitoring data sources and detecting changes without human intervention. The automated data change detection system continuously queries data sources, compares new data with existing records, and updates the database autonomously, eliminating the need for manual verification while maintaining high data accuracy
Solution Approach 2:
The patent replaces the mechanical manual verification process with an automated electronic system. The system uses computer-based algorithms to monitor data sources, detect changes, and update records, substituting human labor with automated computational processes that are both faster and more reliable
2Loss of time
If data is updated frequently from multiple sources to maintain current information, then data freshness is improved, but data inconsistency and errors increase
Solution Approach 1:
The system implements feedback mechanisms by continuously monitoring data sources and comparing new data against existing records. When changes are detected, the system processes the updates through a controlled workflow that validates data consistency, ensuring that frequent updates do not compromise data reliability
Solution Approach 2:
The patent applies preliminary action by establishing data validation rules and consistency checks before updates are committed to the database. The system prepares and verifies data changes in advance, ensuring that only consistent and error-free updates are applied, thus preventing data inconsistency before it occurs
3Productivity
If machine learning models are trained with historical data to predict personal information, then prediction capability is improved, but model accuracy decreases due to temporal variations in data
Solution Approach 1:
The system applies dynamics by transitioning from static historical data to dynamic real-time data. The automated monitoring system continuously captures current data states, allowing machine learning models to be trained on up-to-date information that reflects current conditions, thereby maintaining prediction accuracy despite temporal variations
Solution Approach 2:
The patent uses preliminary action by proactively collecting and storing current data before it becomes historical. The system continuously monitors and updates the database with the latest information, ensuring that training data is always current and relevant, which improves model accuracy compared to using stale historical data
Data Source
AI summary
To train models, training data is needed. As personal data changes over time, the training data can get stale, obviating its usefulness in training the model. Embodiments deal with this by developing a database with a running log specifying how each person's data changes at the time. When data is ingested, it may not be normalized. To deal with this, embodiments clean the data to ensure the ingested data fields are normalized. Finally, the various tasks needed to train the model and solve for accuracy of personal data can quickly become cumbersome to a computing device. They can conflict with one another and compete inefficiently for computing resources, such as processor power and memory capacity. To deal with these issues, a scheduler is employed to queue the various tasks involved.