Load ReutersCorn$-$train.arff and apply the StringToWordVector filter with case$-$folding $($i$.$e$.,$ set lowerCaseTokens to true$).$

On the Preprocess panel, delete all but the two attributes, corn and maize, plus the class. $($Click the button that selects all attributes; then find the three you want and deselect them. Delete all the rest.$)$ Save the resulting dataset $($use a different name without overwriting the original file$).$

Load ReutersCorn$-$test.arff, process it in the same way, and save it $($use a different name without overwriting the original file$).$

In the Classify panel, apply Naive Bayes to the training set and test it on the test set. What is the

Overall accuracy?

$[$ Select $]$

Number of errors on the minority class?

$[$ Select $]$

Number of errors on the majority class?

$[$ Select $]$

ROC area?

$[$ Select $]$