Prediction Device Using Multiple Interpolation Methods for Time Series Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing prediction models face degradation in accuracy when interpolating values from time series data that is not stably acquired, particularly in environments where data is intermittently collected.
Innovation Solution
A prediction device that receives time series data and uses multiple interpolation methods to generate interpolated values, identifying a range between the maximum and minimum values of these interpolated values as the prediction result, thereby suppressing bias from specific interpolation methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple interpolation methods are used to generate interpolated values, then prediction accuracy is improved by suppressing bias from specific methods, but device complexity increases due to multiple processing methods
Solution Approach 1:
The patent segments the interpolation process into multiple independent methods (first interpolation method, second interpolation method, etc.), where each method processes the time series data separately to generate its own interpolated values. This segmentation allows comparison of results from different methods while maintaining modular processing that can be independently implemented and optimized.
Solution Approach 2:
The patent merges the results from multiple interpolation methods by identifying a common range (intersection) that all methods agree upon. This merging process combines the strengths of different interpolation approaches while filtering out method-specific biases, producing a more reliable prediction range that reflects consensus across multiple processing paths.
2Measurement precision
If data acquisition frequency is increased to reduce interpolation needs, then prediction accuracy is improved, but loss of time increases due to more frequent measurements
Solution Approach 1:
The patent performs preliminary actions by generating multiple interpolated values using different methods before final prediction. This preliminary processing creates a set of candidate values that can be evaluated and compared, allowing the system to select the most reliable prediction range without requiring additional real-time data acquisition, thus saving time while maintaining accuracy.
Solution Approach 2:
The patent creates copies of the interpolation process through multiple different methods, where each method generates its own set of interpolated values as a copy of what the true values might be. By comparing these copies and finding their common range, the system achieves accurate predictions without needing to acquire more original data points, thereby avoiding additional time loss.
Data Source
AI summary
An interface receives time series data including multiple observed values of an observed parameter that are acquired at different time points for obtaining physiological information of a subject. A processor inputs the time series data into a prediction model to output, as a prediction result, a range within which values interpolating the multiple observed values may fall. The processor generates at least one first interpolated value by interpolating the multiple observed values with a first method. The processor generates at least one second interpolated value by interpolating the multiple observed values with a second method that is different from the first method. The processor identifies, as the prediction result, a range defined between a maximum value and a minimum value among the first interpolated value and the second interpolated value.