In text mining it seems obvious that we should use all the data we can get our hands on for use in drawing conclusions. The temptation is always to use the broadest possible query to select the data set, because we don’t want to miss anything that might be important. The problem with such an ...
