When you have large numbers of observations in your dataset and all the classes to be predicted are sufficiently represented by the data points, then deleting missing observations would not introduce bias or disproportionality of output classes.
In the housing.dat dataset, we saw from the summary statistics that the dataset has two columns, ptratio and rad, with missing values.
The na.omit() function lets you remove all the missing values from all the columns of your dataset, whereas the complete.cases() function lets you remove the missing values from some particular column/columns.
Sometimes, particular variable/variables might have more missing values than the rest of the variables in the dataset. Then it is better to remove ...