Linear discriminant analysis is one of the oldest statistical models for classification problems, and along with logistic regression is one of the most widely-known methods for classifying data. Although it has become less widely used as more sophisticated classification methods have been developed, it remains a baseline methodology for comparison of new approaches. Its strengths include its speed in training and ease of application in classifying new instances. While significance tests employed in discriminant analysis depend on often unrealistic assumptions such as normality, in situations where large amounts of data are involved and significance tests are of less value, it often remains competitive with more modern methods.
Like your visualization? Why not deploy it? For more information, see Deploy a model.