Predictive Model Stream Prioritization for Edge Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The proliferation of data from IoT sensors and other sources in value chain networks overwhelms traditional centralized data collection methods, leading to complexity and inefficiencies in data transmission and automated decision-making.
Innovation Solution
A method for processing queries in a distributed database using edge devices, where queries are stored on a dynamic ledger, generating approximate responses based on summary data, and transmitting these responses, with the option to use a blockchain for data storage and a neural network for probability distribution modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If centralized data collection methods are used to gather data from IoT sensors and other sources, then complete data collection is achieved, but network overhead and complexity increase significantly
Solution Approach 1:
The patent divides the centralized data collection system into distributed edge devices that operate autonomously. Each edge device segments the data collection and processing tasks, maintaining local data streams and making decisions locally, thereby reducing the burden on centralized collection while preserving data completeness through distributed aggregation.
Solution Approach 2:
The patent introduces a new dimension of processing by implementing predictive models that generate future data values. This transforms the data collection approach from purely reactive (collecting existing data) to proactive (predicting future states), reducing the need for continuous centralized data gathering while maintaining reliability.
2Measurement precision
If all predictive model data streams are processed in detail, then prediction accuracy is improved, but computational resources and time are overwhelmed
Solution Approach 1:
The patent implements prioritization mechanisms that process only the most critical data streams in full detail while using summary statistics for less critical streams. This partial action approach maintains prediction accuracy for high-priority models while preserving computational resources, effectively resolving the contradiction between precision and productivity.
Solution Approach 2:
The patent applies different processing qualities to different data streams based on their priority and importance. High-priority predictive models receive detailed processing with full model parameters, while lower-priority streams receive summarized processing, optimizing the balance between accuracy and computational efficiency across the system.
3Measurement precision
If detailed predictive model parameters are transmitted and processed, then model accuracy is maintained, but data transmission volume and processing time increase
Solution Approach 1:
The patent pre-processes predictive model data by generating summary statistics and prioritizing data streams before transmission. This preliminary action reduces the volume of data that needs detailed processing and transmission, while ensuring that high-priority models receive their full parameters when needed, thereby reducing overall processing time without sacrificing accuracy for critical models.
Data Source
AI summary
A method for prioritizing predictive model data streams includes receiving, by a device, a plurality of predictive model data streams. Each predictive model data stream includes a set of model parameters for a corresponding predictive model. Each predictive model is trained to predict future data values of a data source. The method includes prioritizing, by the device, each of the plurality of predictive model data streams. The method includes selecting at least one of the predictive model data streams based on a corresponding priority. The method includes parameterizing, by the device, a predictive model using the set of model parameters included in the selected at least one predictive model data stream. The method includes predicting, by the device, the future data values of the data source using the parameterized predictive model.