> Does anyone have a source to recommend that
> documents a method for developing a stoplist
> for a specific corpus rather than a generic
> one for, say, all English texts?
I have sometimes executed a word-frequency program, then collected the
100 (or whatever) most common words in that text, and used that collection
as the stoplist (say for an index of the text). Good luck with your work.
--Eric Johnson
johnsone@jupiter.dsu.edu
http://www.dsu.edu/~johnsone/