Clustering Device Using Evaluation Scores for Optimal Cluster Count
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing clustering algorithms face inefficiencies in handling new data points, either sacrificing accuracy for speed or vice versa, particularly when determining the optimal number of clusters and updating cluster assignments.
Innovation Solution
A clustering device and method that calculates an evaluation score based on internal compactness and external separation to determine the optimal number of clusters and efficiently assign new data points to existing clusters, ensuring high accuracy and speed through a batch process and update process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the clustering algorithm determines the optimum number of clusters using all element data including new element data and performs clustering with the optimum number of clusters, then the classification accuracy is improved, but the computation load increases and response speed decreases
Solution Approach 1:
The patent pre-calculates and stores evaluation scores for different numbers of clusters using batch processing before new data arrives. When new element data is added, the system retrieves pre-computed evaluation scores and performs only minimal updates instead of complete re-clustering, thus maintaining high accuracy while achieving fast response
Solution Approach 2:
The system dynamically switches between two modes: batch process mode for initial clustering with optimum cluster determination, and update process mode for rapid incorporation of new data. This dynamic approach allows the system to optimize for accuracy when needed and for speed when new data arrives
2Productivity
If the clustering algorithm assigns new element data to the cluster with minimum distance to the new element data, then the response speed is improved, but the classification accuracy may decrease
Solution Approach 1:
The system pre-computes and stores evaluation scores for different cluster configurations before new data arrives. When new element data is added, it uses these pre-computed scores to guide the assignment decision rather than relying solely on distance metrics, ensuring both speed and accuracy
3Device complexity
If the number of clusters is set in advance using k-means clustering algorithm, then the computation is simplified, but the determination of optimum number of clusters requires repeated clustering with different numbers of clusters
Solution Approach 1:
The patent pre-calculates evaluation scores for multiple possible numbers of clusters in advance using batch processing. These pre-computed scores are stored and can be quickly retrieved when needed, eliminating the need for repeated trial-and-error clustering experiments and significantly reducing the time required to determine the optimum number of clusters
Data Source
AI summary
A clustering device includes: an evaluation score calculation section configured to calculate an evaluation score or evaluation scores for evaluating a classification result; a batch process section configured to classify multiple element data into clusters with an optimum number of clusters, based on the evaluation scores respectively obtained for different number of clusters by assigning each of the multiple element data to one of the clusters; an update process section configured to assign newly added element data to a cluster that is closest to the newly added element data among the clusters into which the multiple element data are classified by the batch process section; and a determination section configured to determine validity of a classification result after assigning the newly added element data to the cluster, based on the evaluation score obtained by assigning the newly added element data to the cluster by the update process section.