Two-Stage Sparse Modeling for Accurate Variable Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sparse modeling techniques face challenges in accurately extracting explanatory variables with a high degree of influence on objective variables, particularly when explanatory variables are similar or have multicollinearity, leading to potential omissions in identifying key factors.
Innovation Solution
The information processing apparatus employs a two-stage sparse modeling approach using first and second regression equations to extract first and second explanatory variables, respectively, with specific units for updating regression coefficients, calculating regression errors, and determining convergence conditions to ensure accurate extraction without omission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing sparse modeling is used to extract explanatory variables, then the extraction process is simple, but explanatory variables with high influence cannot be extracted without omission when variables are similar or have multicollinearity
Solution Approach 1:
The patent divides the single sparse modeling process into two sequential stages: first sparse modeling to extract initial explanatory variables, and second sparse modeling to extract additional explanatory variables from the first results. This segmentation allows comprehensive extraction of all high-influence variables while managing complexity through structured multi-stage processing.
Solution Approach 2:
The patent performs preliminary extraction of explanatory variables using first sparse modeling before conducting the second sparse modeling. This preliminary action ensures that initial high-influence variables are identified first, which then serve as input for finding additional variables, preventing omissions while maintaining systematic complexity control.
2Measurement precision
If existing sparse modeling is used, then the processing is fast, but it cannot appropriately extract explanatory variables when they are similar to each other
Solution Approach 1:
The patent segments the variable extraction task into two phases: first sparse modeling handles initial extraction, and second sparse modeling handles remaining similar variables. This segmentation improves accuracy for similar variables by giving them dedicated processing attention while limiting time loss through efficient sequential execution rather than exhaustive single-stage processing.
Solution Approach 2:
The patent performs sparse modeling twice, which is more than the single application in existing methods. This excessive action ensures that similar explanatory variables are thoroughly extracted by applying the same effective process multiple times, improving completeness while accepting reasonable additional time investment for accurate results.
3Measurement precision
If two-stage sparse modeling is implemented, then extraction accuracy is improved, but the complexity of the modeling process increases
Solution Approach 1:
The patent segments the complex extraction task into two manageable sparse modeling stages, each with its own regression equation. This segmentation improves completeness by ensuring all high-influence variables are captured while controlling processing complexity through modular, repeatable units rather than a single complex monolithic process.
Solution Approach 2:
The patent uses the same sparse modeling methodology for both first and second regression equations, making the process universal and reusable. This multi-functionality approach improves extraction completeness by applying the proven method twice, while keeping complexity manageable through consistency and reusability of the same core algorithm.
Data Source
AI summary
An information processing apparatus comprising processing circuitry. The processing circuitry is configured to acquire objective variables and explanatory variables which are regression analysis targets, extract a plurality of first explanatory variables having a high degree of influence on the objective variable from among the explanatory variables by sparse modeling using a first regression equation, and extract a second explanatory variable having a high degree of influence on the plurality of first explanatory variables by sparse modeling using a second regression equation.