Using Principal Component Analysis and Linear Discriminant Analysis as Dimensional Reduction Techniques رسالة ماجستير

Abstract

principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) will be used as dimensional reduction techniques. In particular, PCA will be compared with other dimensional reduction technique which is Linear Discriminant Analysis(LDA), these two methods and others are used to reduce the number of random variables and obtaining a set of principal variables that retains a large percentage of the total variation. These two techniques will be applied on a dataset and explored and compared. The comparison will be done between the two mentioned methods. We have relied on the number of components after reduction to give the best proportion of variance retained, so the total variance after reduction with the same number of components will determine the best method. Recommendation will be made and the results will be presented