When performing a separation of test results, coping with enormous high-dimensional data sets is necessary but problematic. The input of high-dimensional data, in which not a few elements are irrelevant or less relevant than others, usually lead to inadequate results. It is therefore useful to consult methods, which classify the individual dimensions of the data volumes according to their relevance. In this paper, we present the Principal Component Analysis (PCA) and a Self-developed non-linear Data Analysis (SEDA), used on a complete data collection, as classification methods. Both analyzes are clarified using the same example.