Please read and agree to the terms and conditions of this site.
@raamana_ 4 years ago
tips to review #MachineLearning papers, esp. in #medicine & #neuroscience: -ask for a flowchart from raw data to final metrics! -don't assume CV (like repeated holdout) and advanced metrics (AUC) are implemented correctly. If you don't follow it drunk, they're likely wrong!
@LauraBBalzer 5 years ago
Last ? from #DataScience 4 Public Health course- Who would you trust: a blackbox machine learning algorithm with 100% CV-prediction accuracy (AUROC=1) built with representative data OR a human medical provider (accuracy<100%; AUROC<1)? #epiTwitter #MachineLearning #SuperLearner