When you reclassify a message, POPFile will add the words in that message to a list of words that is associated with the bucket that you reclassified the message to. Thus for each bucket there is a list of associated words. All the bucket lists together form the corpus.
The corpus eventually determines how each message is categorized. It is therefore the most valuable part of your POPFile installation, since it contains all of POPFile's knowledge about the way you like to categorize your mail. You are encouraged to back up your corpus at regular intervals.
Should you find anything in the documentation that is incomplete, unclear, outdated or just plain wrong, please let us know and leave a note in the Documentation Forum.