What is linear Discriminant analysis?
Linear Discriminant Analysis (LDA) is a type of linear combination, a mathematical process that uses different elements of data and applies functions to that group to analyze multiple classes of objects or elements separately. Linear discriminant analysis, derived from Fisher's linear discriminant analysis, can be useful in areas such as image recognition and predictive analysis in marketing.
The basic idea of linear combinations goes back to the 1960s with the Altman Z-scores for bankruptcy and other predictive constructs. Linear discriminant analysis now helps to present data for more than two classes when logical regression is not enough. Linear discriminant analysis uses the mean for each class and takes variants into account to make predictions assuming a Gaussian distribution. It is one of several types of algorithms that is part of making competitive machine learning models.