ml-specialty video for a governmental dataset includes, but not limited to, the following columns: “ID”, “Gender”, “Age”, “Income_month”, “debt”, “Income_year”.
A governmental dataset includes, but not limited to, the following columns: “ID”, “Gender”, “Age”, “Income_month”, “debt”, “Income_year”. The target label is whether this specific citizen would be a criminal or not. A machine learning model should be built for this prediction using SageMaker’s XGBoost algorithm. The “Income_month” and “Income_year” are found to have a correlation of 0.97. What is the recommended action to be taken for such columns?