Computer scienceData scienceMachine learningEnsemble learning

Ensemble methods

Dealing with a noisy dataset

Report a typo

Imagine that you're trying to build an ML model on a dataset containing some outliers and noisy examples. You don't have time to carefully preprocess the data and remove all the noise, so you'll just use it as it is.

Which model would be a good choice for such a tricky dataset?

Select one option from the list

A bagged ensemble, e.g. Random Forest.

A boosted ensemble such as AdaBoost.

I will not use any ensemble: in the presence of outliers, individual model will work better.

___

Create a free account to access the full topic