A Clustering Algorithm Based on Minimum Spanning Tree and Density

A Minimum spanning tree (MST) clustering algorithm based on density is proposed under the background of no effective support on the dataset with noises. Split and merge stages are employed for the proposed clustering algorithm. Furthermore, density estimation method is designed for split stage, and...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	2019 IEEE 4th International Conference on Big Data Analytics (ICBDA) s. 1 - 4
Hlavní autoři:	Chen, Jiayao, Lu, Jing
Médium:	Konferenční příspěvek
Jazyk:	angličtina
Vydáno:	IEEE 01.03.2019
Témata:	Classification algorithms Clustering algorithms component formatting Heuristic algorithms Image edge detection insert (keywords) Machine learning algorithms Partitioning algorithms Shape style styling
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	A Minimum spanning tree (MST) clustering algorithm based on density is proposed under the background of no effective support on the dataset with noises. Split and merge stages are employed for the proposed clustering algorithm. Furthermore, density estimation method is designed for split stage, and maximal connected sub graph is employed for merge stage. The experiments are performed on synthetic and real datasets, and clustering results show the proposed clustering algorithm can detect clusters with noises. Moreover, clusters with complex shapes can be recognized as well and it demonstrates the proposed clustering algorithm is robust to noises.
DOI:	10.1109/ICBDA.2019.8713247