Linear Discriminant Analysis (LDA)

Linear discriminant analysis (LDA) and the Fisher linear discriminant method is used to find the statistics and to learn specific features or separate two or more classes of objects or events a linear combination of machines. The resulting combination can be used as linear classifier, or more commonly, in the future to reduce the peacekeeping category.

LAD is closely related to ANOVA (analysis of variance) and regression analysis, which also tried to express a variable for other functions or a linear combination of measurements. In the other two methods, however, the dependent variable is a numerical quantity, and the LAD is an absolute variable (ie, class label). Logistic regression and probability of return is more similar to the LAD, because they also shows a classification variable. These other methods in the application, preferably it is unreasonable to assume that, since the variable is normally distributed, which is a basic assumption of the LAD method.


LAD is also closely related to principal component analysis (PCA) and factor analysis in a linear combination of these two variables to see if the best interpretation of data. The Department clear attempt to model the class differences between the data. On the other hand, the Permanent Court of Arbitration does not consider any class differences, and factor analysis, based on the basis of functional differences, rather than the same combination. Discriminant analysis is also different from the factor analysis, this is not an interdependent technologies: an independent variable and the difference between the dependent variable (also known as the standard variables) must be made.

LAD work, each observed variable measurement is achieved by a continuous number. When combined with an absolute independent variables treatment, equivalent technical Discriminant Correspondence Analysis