Suppose that there is a color column (feature) in a data frame with values in ("Blue", "Green", "Black","Red"). What is the best way to prepare this column for modeling?NormalizationReplace the values with numbers (1,2,3,4)O Label encodingOne-hot encoding After calculating Mean Square Error (MSE) and Mean Absolute Error (MAE) for a linear model, weobserve that the value of MSE is much larger than the value of MAE. What are the possible reasons?Both A and BC. The data is not linearO B. There are outliers in the datasetO A. MAE is in y unit but MSE is squared

In this question we have been given MCQ questions related to Machine Learning (ML) models and we need to choose the correct answer for each given question.Solution 1:- Option (3) is correct => label encoding. Explanation:- While modeling some of the models does not work with categorical values. So for handling categorical values perform label encoding where we use factor to convert each categorical data to numeric data. Option (1) is incorrect as normalization is done in case of features have so much difference in values so here we do not need normalization. Option (2) is incorrect as it is not required every time to replace these values we also need to convert it to numeric type which is done by label encoding. Option (4) is incorrect as one-hot encoding is not required in this case.Solution (2):- Option (1) is correct => Both A and B Explanation:- There are two possible reasons of greater value of Mean Square Error (MSE) as compared to Mean Absolute Error (MAE), first one is obvious which is MSE is squared that is why it has higher value. Second reason for greater value of MSE that can be outliers in the dataset which results in very large value of MSE as compared to MAE. Option (2) is incorrect as it is not the valid reason about larger value of MSE as compared to MAE. Option (3) is incorrect as this is not the only possible reason for greater value of MSE, option (A) is also correct. Option (4) is incorrect as this is not the only possible reason for greater value of MSE, option (B) is also correct.