A Linear Transformation of Feature Extraction for High Dimensional Datasets Using K-Means clustering

Abstract

Data mining employs a variety of traditional statistical methods such as cluster analysis, discriminate analysis, logistic regression, and time series forecasting. Due to the mega high dimensionality nature of datasets, data dimension reduction has drawn special attention for such type of data analysis. Feature extraction can be viewed as preprocessing step which removes distracting variance from the datasets so that clustering, classifiers can estimators perform better. In this paper principal component analysis, a linear transformation is used for dimensionality reduction and clustering with K-Means algorithm is applied and shows the results.

Keyword: Principal component analysis, dimensional reduction, k-means clustering.