Overlap Matrix Weighting for Data Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computing systems with limited resources face challenges in performing computationally intensive algorithms due to the large number of calculations required for grouping data items, leading to a 'computationally explosive' situation where the desired algorithm cannot be completed within a reasonable time frame.
Innovation Solution
The method generates overlap matrices from data attributes, combines them with weights to create weighted overlap matrices, and then generates similarity matrices, which are subjected to clustering analysis to identify commonly occurring groupings, thereby reducing the computational load and avoiding direct comparison of all data sets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct comparison of all data sets is performed to identify groupings, then comprehensive grouping analysis is achieved, but computational load becomes explosive and exceeds available computing resources
Solution Approach 1:
The patent segments the computational task by dividing it into multiple iterations, where each iteration processes a subset of data sets rather than all data sets simultaneously. This segmentation reduces the computational load per iteration while progressively building toward comprehensive grouping analysis across all data sets over multiple passes.
Solution Approach 2:
The patent applies preliminary action by performing clustering analysis on a subset of data sets before processing all data sets. The groupings identified in preliminary iterations serve as initial conditions or constraints for subsequent iterations, reducing the search space and computational requirements for comprehensive analysis.
2Measurement precision
If multiple attributes are analyzed to improve grouping accuracy, then measurement precision increases, but the number of calculations increases exponentially
Solution Approach 1:
The patent employs dynamics by adaptively adjusting the number of data sets processed in each iteration based on computational resources available and progress made. The system dynamically modifies processing parameters such as subset size and iteration count to balance grouping accuracy with computational feasibility, rather than using a fixed approach.
Solution Approach 2:
The patent changes parameters by varying the number of data sets included in each clustering iteration and adjusting weighting factors for different attributes. By modifying these parameters across iterations, the system achieves comprehensive multi-attribute analysis while controlling computational complexity through progressive refinement rather than simultaneous processing of all parameters.
Data Source
AI summary
A method including generating a first overlap matrix from a first attribute having first measurements of data items, and generating a second overlap matrix from a second attribute having second measurements of the data items. Samples of weights are generated, each of the samples of weights including a corresponding first weight for the first overlap matrix and a corresponding second weight for the second overlap matrix. For each of the samples of weights, the first overlap matrix is combined with the corresponding first weight and the second overlap matrix is combined with the corresponding second weight. Similarity matrices are generated by combining, for each of the samples of weights, a weighted first overlap matrix with a weighted second overlap matrix. A cluster analysis is performed on the similarity matrices to generate groupings the similarity matrices. The groupings represent groups of the data items. A selected grouping is chosen.