Grid Clustering Algorithm with Simple Leaping Search Technique

Data mining is a critical data analysis technique for extracting hidden information from large databases for business or industrial applications. As the size of organizational databases increase, finding information and knowledge efficiently is essential. In the past, numerous clustering algorithms...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2012 International Symposium on Computer, Consumer and Control s. 938 - 941
Hlavní autoři: Tsai, Cheng-Fa, Zhang, Jun-Hao
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.06.2012
Témata:
Algorithm design and analysis
Clustering algorithms
data clustering
Data mining
large database
Noise
Partitioning algorithms
Spatial databases
ISBN:9781467307673, 146730767X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Data mining is a critical data analysis technique for extracting hidden information from large databases for business or industrial applications. As the size of organizational databases increase, finding information and knowledge efficiently is essential. In the past, numerous clustering algorithms based on grid-clustering schemes have been proposed. This study proposes, simple-leaping search (SLS), a new grid-based clustering algorithm that partitions the space by the number of grids. It then sequentially searches odd columns of all grids according to the minimal point set at each grid. Based on whether the grid is useful or useless, different neighbor grids are searched. Experimental results show that the SLS clustering algorithm performs better than other clustering algorithms such as DBSCAN, IDBSCAN and GOD-CS.
ISBN:9781467307673
146730767X
DOI:10.1109/IS3C.2012.244