Clustering Algorithm Recalculates Initial Conditions for Data Distribution Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data analysis methods fail to update cluster structures accurately when point data is added, updated, or deleted, as they do not recalculate initial condition values based on changes in data distribution, leading to incorrect inclusion or exclusion of data in clusters.
Innovation Solution
A data analysis device and method that recalculates initial condition values and updates cluster structures by determining whether point data is a core or noise using a predetermined algorithm, allowing for re-verification of point data and updating the cluster structure accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sequential clustering is performed without recalculating initial condition values, then processing speed is maintained, but clustering accuracy deteriorates when data distribution changes
Solution Approach 1:
The patent performs preliminary recalculation of initial condition values (EPS and MinPts) when data distribution changes are detected, such as when new point data is added or existing data is updated. This preliminary action ensures that the clustering parameters are current before performing the actual clustering operation, thereby maintaining accuracy without requiring continuous full recalculation
Solution Approach 2:
Instead of recalculating all clustering parameters for the entire dataset every time, the patent applies partial recalculation only for affected regions or parameters. For example, when new point data is added, only the EPS value may be recalculated based on the new data distribution, while MinPts remains unchanged, thus reducing computational overhead while maintaining necessary accuracy
2Adaptability or versatility
If initial condition values are fixed, then processing complexity is reduced, but adaptability to data distribution changes deteriorates
Solution Approach 1:
The patent implements dynamic initial condition values that automatically adjust based on data distribution changes. The EPS parameter is dynamically recalculated using density-based methods when new point data arrives or existing data is updated, while MinPts remains a static threshold. This dynamic adaptation allows the clustering algorithm to respond to changing data patterns without requiring complex manual reconfiguration
Solution Approach 2:
The system incorporates feedback mechanisms where the clustering process monitors data distribution changes and triggers recalculation of initial condition values when necessary. When new point data is added or updated, the system detects the change in data distribution and automatically adjusts EPS and MinPts accordingly, creating a closed-loop system that adapts to changing conditions while maintaining manageable complexity through automated decision-making
3Measurement precision
If neighborhood point data only is considered, then processing speed is improved, but measurement precision of data distribution changes deteriorates
Solution Approach 1:
The patent segments the data processing into two distinct phases: a fast preliminary phase that processes only neighborhood point data to maintain speed, and a detailed verification phase that performs comprehensive data distribution analysis when changes are detected. This segmentation allows the system to maintain high processing speed for routine operations while ensuring measurement precision when accuracy is critical
Solution Approach 2:
The system performs preliminary clustering using only neighborhood point data to quickly identify potential clusters and maintain processing speed. After this preliminary action, a verification step checks whether the clustering results are consistent with the overall data distribution. If inconsistencies are found, a more comprehensive analysis is performed, thus maintaining speed for most operations while ensuring precision when needed
Data Source
AI summary
A data analysis device is configured to include an arithmetic operation device that determines whether or not each point data in the point data group of the existing clusters is a core of the cluster using a predetermined algorithm and calculates a predetermined re-verification determination data, recalculates an initial condition value of the clustering based on the re-verification determination data, specifies re-verification point data which is the point data that becomes no longer the core or becomes a new core by adding the point data, from the point data group to which the addition was made, based on the updated initial condition value obtained by the recalculation, and performs predetermined clustering processing in accordance with the re-verification point data.