DRAG DROP You are producing a multiple linear regression model in Azure Machine Learning Studio. Several independent variables are highly correlated. You need to select appropriate methods for conducting effective feature engineering on all the data. Which three actions should you perform in sequence? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order. 
Actually, it's better to start with Remove duplicate rows because duplicates can throw off correlation measures and model training. Then Compute linear correlation helps you spot multicollinearity among your features. Last, Use the Filter Based Feature Selection module lets you filter out less useful predictors. Some might pick "Test hypothesis using t-Test" but that's more for comparing means, not selecting features. Anyone see it differently?
- Action 1: Remove duplicate rows
- Action 2: Compute linear correlation
- Action 3: Use the Filter Based Feature Selection module
I don't think "Remove duplicate rows" should be first here. I'd start with "Compute linear correlation," then "Test the hypothesis using t-Test," and last "Use the Filter Based Feature Selection module." I think removing duplicates isn't always needed up front. Disagree?
