Linear discriminant analysis (LDA) - how to use it as a classifier

preview_player
Показать описание

In this second video about LDA we will see how LDA can be used as a machine learning technique. We will first discuss the basics of LDA, then (6:09), we will see how we can find a good cutoff value for classification, how to calculate the sensitivity and specificity (08:00), ROC curve (11:56), probabilities (13:50) and how we can make predictions based on more than two groups (18:56) by using the Iris data set.

Note that I incorrectly say 30.5 at (18:17 and 18:38). I should have said, according to the text, 31.5.
Рекомендации по теме
Комментарии
Автор

Great video! Helped me to understand what LDA is quickly. Keep it up with the videos!

uberheropatapon
Автор

Another good video. I wish you would elaborate more on probabilities, in particular when more than two categories come into play. How probabilities are reported when we have three categories? Furthermore, it would be good if this video was accompanied by some explanations about QDA

abbasatashdehghan
Автор

Excellent video. How do you get 0.132 and 0.062 for a score of 30.8? Please guide me. Thank you

anuradhasriram
Автор

Can you please let me know how to normalize weights after calculating eigen vector lda score function

rzcjtou
Автор

Are those values 0.7, 0.11 are eigen vectors?

sharp_guy