Segmented Data Forecasting via Clustered Model Combination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex systems face challenges in reliably and efficiently determining future behavior at a granular level, as distinguishing between portions of input data that may behave differently over time is inefficient, reducing system effectiveness and resource utilization.
Innovation Solution
The system automatically segments input data using clustering metrics, associates each subset with a supervised learning model optimized for its characteristics, and combines these models to generate forecasts with higher granularity and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single model is used to process all input data, then the system structure is simple, but the forecast accuracy and granularity are reduced
Solution Approach 1:
The patent divides the input data into multiple segments based on clustering metrics, where each segment represents a subset of data with similar characteristics. Separate supervised learning models are then trained for each segment, allowing each model to specialize in predicting specific patterns within its segment. This segmentation approach enables higher forecast accuracy and granularity while maintaining manageable system complexity through modular architecture.
2Measurement precision
If data is segmented into multiple subsets with separate models, then forecast granularity and accuracy improve, but computational resources and processing time increase
Solution Approach 1:
The patent performs preliminary clustering of data into segments before model training, organizing data into distinct subsets based on their characteristics. This preliminary organization enables efficient parallel processing of multiple models, where each model processes only its assigned segment. By preparing data structures in advance and using parallel computation, the system achieves high predictive accuracy across multiple segments while optimizing computational resource utilization through distributed processing.
3Productivity
If the system processes data at a single level of granularity, then processing is efficient, but the ability to distinguish between different portions of input data is reduced
Solution Approach 1:
The patent applies segmentation at multiple levels of granularity by first clustering data into broad segments, then potentially further subdividing these segments into finer subsets. Each segment or subset can be processed by dedicated models that are optimized for its specific characteristics. This multi-level segmentation enables the system to maintain processing efficiency at each level while progressively improving the ability to distinguish and analyze different portions of input data with increasing precision.
Data Source
AI summary
Segmenting data and forecasting by a combination of models trained on segmented data is provided. A system compares, with a first model, values of timestamps corresponding to data points to determine a time series dependency between the data points. The system generates, with the first model and based on the time series dependency, a first cluster with first data points and a second cluster with second data points. The system allocates, by a controller, a second model to the first cluster, and a third model to the second cluster. The system trains the second model based on the time series dependency and the first data points. The system trains the third model based on the time series dependency and the second data points. The system generates a fourth model based on a combination of the second trained model and the third trained model.