[spambayes-dev] Tricky false positive: US states

[Neil Schemenauer]

On Sat, Oct 04, 2003 at 06:11:45PM -0400, Tim Peters wrote:
> That's what happened here.  The rub is that getting the same judgment from
> 100 consultants isn't *really* more reliable than getting it from one
> consultant unless the consultants are independent -- if they are
> independent, very high confidence is fully justified.  In this case, the
> consultants are all related, biased in the same direction for a reason.

It occurs to me that if Spambayes could understand some degree of
dependence between tokens, multi-user databases might work much
better.  For example, one person might like Amazon pseudospam while
another may consider it outright spam.

  Neil