Linear diagnostic evaluation and Fischer linear diagnostics are common methods used, including in machine learning and pattern recognition, to find linear combinations of properties that Best separate two or more classes of objects. Y>is closely related to variance analysis and regression analysis, which attempt to express an independent variable as a linear combination of other features. This independent variable in N-forms the label of a class. It is also closely related to the analysis of the main phenomena. Because both methods look for a linear combination of variables that best describe the data. It also tries to model differences between different classes of data. CDA is used when the sizes of observations are continuous values.