DENSVAR Clustering Algorithm for Varying Density Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing clustering algorithms are inefficient in identifying clusters with varying densities, requiring multiple executions and being computationally costly, which limits their utility for rapid processing and large datasets, especially in applications like infectious disease monitoring and traffic management.
Innovation Solution
The DENSVAR clustering algorithm identifies clusters with varying densities by determining distances between nodes, assigning candidate nodes to clusters based on density parameters, and evaluating these clusters against specified density values, allowing for the identification of clusters with different densities in a single process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing clustering algorithms are executed multiple times to identify clusters with different density levels, then cluster identification accuracy is improved, but processing time and computational complexity increase significantly
Solution Approach 1:
The patent combines multiple clustering operations with different density parameters into a single unified execution. The system processes all density levels simultaneously by maintaining multiple candidate clusters in memory and evaluating them against different density thresholds in one pass through the data, eliminating the need for sequential re-executions
Solution Approach 2:
The system performs preliminary actions by pre-calculating and storing distance information and candidate cluster structures before final density evaluation. By preparing candidate clusters in advance and organizing data structures efficiently, the system enables rapid density assessment across multiple levels without requiring repeated full clustering executions
2Measurement precision
If existing clustering algorithms are executed multiple times to identify clusters with different density levels, then comprehensive cluster detection is improved, but computational cost increases
Solution Approach 1:
The patent merges multiple clustering computations into a single computational pass by maintaining a set of candidate clusters that can be evaluated against different density thresholds simultaneously. This approach processes all density levels in one execution rather than requiring separate computations for each density level
Solution Approach 2:
The system dynamically adjusts density threshold evaluation during a single execution by maintaining candidate clusters at different density levels and selectively validating them based on the specified density parameter. This dynamic approach allows comprehensive cluster detection across varying densities without the fixed computational overhead of multiple static executions
3Adaptability or versatility
If existing clustering algorithms are used for large databases, then cluster analysis capability is maintained, but processing efficiency decreases
Solution Approach 1:
The patent segments the clustering process into candidate cluster identification and density validation phases. By organizing data into candidate clusters based on spatial proximity first, then evaluating density thresholds separately, the system efficiently handles large databases while maintaining comprehensive cluster analysis capability across different density levels
Data Source
AI summary
The present disclosure provides a clustering device configured to detect varying density. The clustering device may receive a set of parameters (e.g., a density parameter specifying a plurality of density values, a size parameter, and a node parameter that includes information associated with a plurality of nodes) for identifying clusters. The clustering device may determine a distance between different pairs of nodes of the plurality of nodes and identify candidate nodes of the plurality of nodes based on the distance determined for the different pairs of nodes and the size parameter. Candidate nodes may be assigned to a candidate cluster that may be evaluated against the density parameter to determine whether a density of the candidate cluster satisfies at least one of the plurality of density values. A cluster may be identified based on whether the candidate cluster satisfies at least one of the plurality of density values.