Comparing the Performances of PCA (Principle Component Analysis) and LDA (Linear Discriminant Analysis) Transformations on PAF (Paroxysmal Atrial Fibrillation) Patient Detection

Linear Discriminant Analysis (LDA) and Principle Component Analysis (PCA) are offered schemes for feature extraction and dimension reduction. They have been used extensively in many applications involving high-dimensional data. In this study, we compared the effectivity of features obtained from PCA and LDA for the diagnosis of Paroxysmal Atrial Fibrillation (PAF) from normal sinus rhythm (NSR) ECG records. Within this framework, a set of features obtained from PCA and LDA were used as an input to the same classification algorithm, which is chosen as the K-Nearest Neighbor (kNN) Algorithm. The obtained results elicit that LDA features have better discrimination capability than those obtained from PCA.