Robust sparse principal component analysis

The Principal Component Analysis (PCA) is the first method of size reduction and data processing when the dataset is of a high-dimensional. Therefore PCA is a widely used method in almost all scientific fields. However, since all the original variables of each principal component in the PCA are linear combination, the interpretation process of the analysis results is often encountered with some difficulties. The approaches proposed for solving these problems are referred to as Sparse Principal Component Analysis (SPCA). However, sparse approaches are not resistant to the fact that there are outliers in the data set, such as in the PCA. In this study, Croux et al. (2013), which combines the advantageous properties of SPCA and Robust Principal Component Analysis (RPCA), the performance of the proposed approach will be examined through two real and three artificial datasets. Preliminary information on PCA, RPCA and SPCA will be given in Section 1 of the study. In the Section 2, a summary of the literature will be included with definitions and explanations of some basic concepts used in the study. The datasets and methods examined in Section 3 will be included, and in Section 4, the methods examined in detail in the previous sections will be applied to real and artificial datasets. In the last section, the results obtained in the study will be discussed.