When estimating these parameters, GraphPad Prism software provides their desired confidence intervals of, for example, 95%. Take a look at it. It is a simple task.

Dear Youssef Lamrani Alaoui

These articles might useful, have a look:

1. https://machinelearningmastery.com/confidence-intervals-for-machine-learning/

2. https://machinelearningmastery.com/report-classifier-performance-confidence-intervals/

3. https://www.machinelearningplus.com/statistics/confidence-interval/

4. https://towardsdatascience.com/how-to-add-confidence-intervals-to-any-model-7bbb9f80fd9c

5. https://vitalflux.com/machine-learning-why-use-confidence-intervals/

6. https://tutorials.one/how-to-calculate-bootstrap-confidence-intervals-for-machine-learning-results-in-python/

Kind Regards

Qamar Ul Islam

Hello Dr Youcef,

I invite you to read these articles:

Article Validating the validation: reanalyzing a large-scale compari...

Article Simple and efficient machine learning frameworks for identif...

Article The Precision-Recall Plot Is More Informative than the ROC P...

Article Using machine learning models to improve stroke risk level c...

Best regards,

Pr Hambaba

Thank you very much for your answers

Quite an interesting read as well would be to transition into bayesian machine learning where each metric comes with its own standard deviation.

https://en.wikipedia.org/wiki/Bayesian_network