Computer Science
5 years ago

Revolutionizing Data Clustering: Faster Updates Without Re-Clustering Needed

Studies in computational intelligence
Improvement of Incremental Hierarchical Clustering Algorithm by Re-insertion
Kakeru Narita, Teruhisa Hochin, Yoshihiro Hayashi, Hiroki Nomiya

Paper Summary

Related papers

Theme background image
Computer Science
3 years ago

New clustering algorithms revolutionize data organization and analysis techniques.

Data Clustering
2 views
Theme background image
Computer Science
4 years ago

New clustering algorithms revolutionize data grouping for more efficient discoveries!

Data Clustering
Multi-label Classification
1 view
Theme background image
Computer Science
5 years ago

Revolutionizing Data Clustering: Faster Updates Without Re-Clustering Needed

Edge Computing
Urban Structure
Data Clustering
1 view

The article presents a new method to update clustering results without re-clustering the entire dataset when new data is added. By considering the multimodality of clusters, points can be moved to different clusters based on their distribution. Additionally, as the number of clusters grows, previously added data points are updated for better accuracy. The new method reduces execution time significantly while maintaining comparable clustering accuracy for certain types of data.