Understanding Linear Discriminant Analysis (LDA)

Understanding Linear Discriminant Analysis (LDA)

Key Points Linear discriminant analysis (LDA) is a supervised learning technique that can be used for classification, dimensionality reduction, feature extraction, clustering, or visualization. It assumes that the data follow a multivariate normal distribution with class-specific parameters and estimates the prior probabilities, the means, and the covariance matrices of each class from the data. It computes the linear discriminant function, a linear combination of features that maximizes the ratio of the between-class variance to the within-class variance. It projects the data onto the linear discriminant. It obtains the LDA score for each data point, which indicates its position and distance from the discriminant function and the likelihood of each data point belonging to each class. It classifies the data points based on the LDA score and the posterior probability of each class. It evaluates the model's performance using various metrics, such as the confusion matrix, the accuracy, th…