文章基本信息

标题：PCAWK : A Hybridized Clustering Algorithm Based on PCA and WK-means for Large Size of Dataset
本地全文：下载
作者：Fatemeh Boobord ; Zalinda Othman ; Azuraliza Abubakar 等
期刊名称：International Journal of Advances in Soft Computing and Its Applications
印刷版ISSN：2074-8523
出版年度：2015
卷号：7
期号：3
出版社：International Center for Scientific Research and Studies
摘要：Real world data usually has a variation of size of dimensionality. The dimensionality needs to be reduced for handling the dimensionality of data. The dimensionality reduction changes the presentation of dimensional data variation to a meaningful presentation. In this paper, a method based on the principle component analysis and WK-means called "PCAWK" is proposed. Firstly, PCA is used to reduce the redundant dimensionality of dataset and then, the WK-means algorithm that is a hybrid of Invasive Weed Optimization (IWO) and the K-means algorithm utilizes the reduced dataset to obtain the optimal clusters. The proposed algorithm is tested on 5 real word instances and the results are compared with the PCAK algorithm. The proposed algorithm generally has better performance in most datasets
关键词：Clustering; K-means; Large Size Data; Metaheuristics; Principle ; Component Analysis