🔒 Please log in to use tool features (fill sample / analyze / AI interpretation / export document)
🔎 Discriminant Analysis (LDA)
Linear discriminant: fit on labeled training data, classify samples / new observations, output a confusion matrix and accuracy
Training data (one row per line: class, feature1, feature2, …)
Samples to classify (optional, one per line: feature1, feature2, …)
Notes
· Linear Discriminant (LDA): estimate each class mean and the pooled within-class covariance Σ, then use the discriminant score δ_k(x)=x′Σ⁻¹μ_k−½μ_k′Σ⁻¹μ_k+ln p_k and take the largest as the predicted class; suitable for numeric features with approximately equal covariance across classes.
· Feature dimension ≥2 recommended; each class ≥3 samples, and total samples > number of features.
About Discriminant Analysis (LDA) Calculator
Free linear discriminant analysis (LDA) calculator: build discriminant functions, classify samples, and view confusion matrix and cross-validation accuracy.
Learn more about this tool →
Principal Component Analysis
Cluster Analysis & EFA
Logistic Regression
Correspondence Analysis
Linear Regression
← Back to all quality tools