Observation Information Processing for Locally Maximum Confidence Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data aggregation techniques face challenges in detecting areas with locally maximum confidence related to association rules, particularly failing to identify areas with locally maximum possibilities, such as those where the likelihood of long-distance taxi rides is not only the highest but also locally significant, due to predetermined mesh divisions leading to the modifiable areal unit problem.
Innovation Solution
An observation information processing apparatus that calculates support and confidence for each mesh, generates adjacent mesh sets, sets a confidence threshold, and repeatedly excludes meshes based on these calculations to detect areas with locally maximum confidence, allowing for the identification of areas with both high and locally maximum possibilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data aggregation is performed by dividing an area into meshes with predetermined division, then the modifiable areal unit problem occurs causing different results depending on division method, but if association rules are used to maximize area detection, then local maximum confidence areas cannot be detected
Solution Approach 1:
The patent divides the search space into hierarchical levels (Level 1: individual meshes, Level 2: groups of adjacent meshes, Level 3: larger aggregated areas). At each level, local maximum confidence areas are detected independently. This segmentation allows multiple local maxima to be found simultaneously without the limitations of predetermined mesh divisions or global optimization approaches.
Solution Approach 2:
Instead of seeking a single global maximum confidence area, the patent performs partial actions by detecting multiple local maximum areas at different hierarchical levels. This excessive detection of local maxima ensures that all significant areas are identified, even if they don't represent the absolute global maximum.
2Productivity
If only the area with the largest confidence is detected, then computation is simplified, but areas with locally maximum confidence are missed
Solution Approach 1:
The detection process is segmented into multiple independent passes, each targeting a different hierarchical level. Level 1 detects local maxima in individual meshes, Level 2 detects local maxima in groups of adjacent meshes, and Level 3 detects larger aggregated areas. This segmentation allows the system to find multiple useful areas without requiring a single complex global optimization algorithm.
Solution Approach 2:
The patent adds a hierarchical dimension to the detection process by operating at multiple levels (individual meshes, groups of meshes, larger aggregated areas). This dimensional approach transforms the problem from finding one global maximum to finding multiple local maxima across different scales, increasing productivity without proportionally increasing complexity.
3Ease of manufacture
If predetermined mesh divisions are used for data aggregation, then the process is simple, but the modifiable areal unit problem causes different results based on division method
Solution Approach 1:
The patent makes the mesh division dynamic by generating different hierarchical levels of meshes adaptively rather than using a single predetermined division. The system creates Level 1 meshes, then Level 2 groups of adjacent meshes, then Level 3 larger aggregated areas, allowing the division structure to adapt to the data distribution and eliminate the modifiable areal unit problem.
Solution Approach 2:
The patent changes the parameter of mesh division from fixed and predetermined to variable and hierarchical. By adjusting the division parameters across different levels (from fine-grained individual meshes to coarse-grained aggregated areas), the system maintains simplicity while ensuring result consistency through multi-level verification.
Data Source
AI summary
An observation information processing apparatus calculates, for each mesh, a support and a confidence. The observation information processing apparatus generates an adjacent mesh set by merging adjacent ones of the meshes. The observation information processing apparatus calculates, based on a support and a confidence of each mesh included in the adjacent mesh set, a confidence for each adjacent mesh, and sets the smallest one of the confidences calculated as a new confidence threshold value. The observation information processing apparatus detects and excludes meshes to be excluded from meshes included in the adjacent mesh set, based on the confidences and supports of the meshes included in the adjacent mesh set and the confidence threshold value.