A Logistic Regression Analysis of LPP and PCA on Classification

In many real-world classification problems the local structure is more important than the global structure and in the dimensionality reduction algorithms such as principle component analysis (PCA) it preserves the global structure of the dataset and ignores the local structure of the dataset, therefore this book introduce the Locality Preserving Projections (LPP) algorithm that is preserving the local structure of the datasets. LPP is a linear projective maps that arise by solving variational problem that optimally preserves the neighborhood structure of the data set. The aims of this book are to compare between PCA and LPP in terms of accuracy, develop appropriate representations of complex data by reducing the dimensions of the data and explain the importance of using LPP with logistic regression. The methodology of this book compared the proposed LPP approach with PCA method on five different data sets using dimensionality reduction toolbox (drtoolbox) in matlab software and evaluation the model using cross validation method and then calculated the performance measures(accuracy, sensitivity, Specificity , precision, f-score and roc curve) of both.