Which steps should the data engineer take to address this issue?
(Choose two.)
Use a linear-based algorithm to train the model.
Apply principal component analysis (PCA).
Remove a portion of highly correlated features from the dataset.
Apply min-max feature scaling to the dataset.
Apply one-hot encoding category-based variables.
Explanations:
A linear-based algorithm may be impacted by multicollinearity, and this does not address the correlation.
PCA reduces dimensionality by transforming correlated features into uncorrelated principal components.
Removing highly correlated features reduces multicollinearity, helping to avoid redundancy in the model.
Min-max scaling changes feature scales but does not reduce feature correlation.
One-hot encoding is used for categorical variables, not for handling correlation among features.
I conclude that the answer is:
Apply principal component analysis (PCA).