Data Fallacy like cherry picking, overfitting, etc.
Mahak from Outgrow
There are so many data fallacies that can happen when we interpret our results or train our models on our dataset. Do you also take care of these data fallacies? Can you pls share how do you get the info about these data fallacies and how do you prevent them?
Well... What do you mean by "prevent"? Data need to be validated before uploading, so it's on Data Analyst side, part of human work inside pure ML
@kpyto By prevent, I meant some data fallacies which led to interpretation and results go in negative direction. Like in data fallacy "false causality" where we falsely assume that two correlated results cause each other. Though its a task of data analyst, but after getting results the analysis should be verified too. I am just asking how to prevent these data fallacies more.