文章基本信息

标题：A Study of Dimensionality Reduction Using Roughset Based K-Means
本地全文：下载
作者：S. Brindha ; Dr. Antony Selvadoss Thanamani
期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2015
卷号：3
期号：8
DOI：10.15680/IJIRCCE.2015. 0308039
出版社：S&S Publications
摘要：Feature Reduction of pattern dimensionality using feature extraction and feature selection belongs to thedata mining. To enhance the robustness of the k-means clustering algorithm and for visualization purpose thedimension reduction techniques may be employed. Randomized Dimensionality reduction is the transformation ofhigh-dimensional data into a significant illustration of reduced dimensionality that corresponds to the fundamentaldimensionality of the data. K-means clustering algorithm often not well for high dimension datasets and errordimensionality reduction, hence, to improve the efficiency, the proposed system apply Roughset theory based k-meanson original data set and obtain a reduced dataset containing possibly uncorrelated variables. In this paper, Roughsettheory for feature selection and K-means based principal component analysis (PCA) for Feature Extraction, non-linearconversion is used for reduce the dimensionality and primary centroid is calculated, then it is applied to K-Meansclustering algorithm.
关键词：K-means; Principal Component Analysis; Roughset; high dimension