Logistic regression : how to use it as a classifier

preview_player
Показать описание

In this second video about logistic regression, we will see how we can use it as a classifier (a machine learning technique) to predict if someone has a disease. The video consists of:
1. Some basics
2. How to classify with a cutoff value (02:15)
3. How to calculate sensitivity and specificity and the ROC curve (08:20)
4. How to validate the model (14:00)
5. Use more than one explanatory variable (15:18)
6. Model selection and predictions (20:39)
Рекомендации по теме
Комментарии
Автор

Very well explained... But you should make an individual ordered playlist for logistic regression and other topics. Makes it easier to follow videos sequentially. Thank You

firstkaransingh
Автор

Nice video. I have a question about using logistic regression with low prevalence (23:25): does NPV decrease due to so many false negatives? On the contrary, we saw the opposite in the example of video about PPV and NPV: FN decrease and FP increase. Maybe we need to use a cutoff value lower than 0.5?

giovannibrufani
Автор

sir can u teach the validation check please
??

MdSajjad-jrsq
Автор

My mri report show on comparison previous mri study's reveal mild to moderate regression inthe thickness rectal neoplasms
This is good or bad for my patients please answer Because tumour size increases

pawantiwari
Автор

thank you very much for this video. The subject is very comprehensive and clearly explained.

PeterJMPuyneers