Title :
An intelligent Weighted Kernel K-Means algorithm for high dimension data
Author :
Kenari, Abdolreza Rasouli ; Maarof, Mohd Aizaini Bin ; Sap, Mohd Noor Bin Md ; Shamsi, Mahboubeh
Author_Institution :
Univ. Teknol. Malaysia, Malaysia
Abstract :
Clustering is a kind of unsupervised classification of objects into groups so that objects from the same cluster are more similar to each other than objects from different clusters. In this paper, we focus on Weighted Kernel K-Means method for its capability to handle nonlinear separability, noise, outliers and high dimensionality in the data. A new WKM algorithm has been proposed and tested on real Rice data. the results exposed by algorithm encourage the use of WKM for the solution of real world problems.
Keywords :
data mining; pattern classification; pattern clustering; unsupervised learning; Rice data; high dimension data; intelligent weighted kernel K-means algorithm; noise; nonlinear separability; object clustering; outliers; unsupervised classification; Atmospheric modeling; Autocorrelation; Classification algorithms; Clustering algorithms; Clustering methods; Data mining; Electrical capacitance tomography; Kernel; Machine learning algorithms; Testing; Classification Accuracy; Clustering; Data Mining; F-Measure; WKM Algorithm; Weighted Kernel K-Means;
Conference_Titel :
Applications of Digital Information and Web Technologies, 2009. ICADIWT '09. Second International Conference on the
Conference_Location :
London
Print_ISBN :
978-1-4244-4456-4
Electronic_ISBN :
978-1-4244-4457-1
DOI :
10.1109/ICADIWT.2009.5273893
