Jeff Abrahamson on 22 Oct 2004 15:00:03 -0000 |
On Fri, Oct 22, 2004 at 10:25:02AM -0400, Tobias DiPasquale wrote: > [30 lines, 141 words, 1152 characters] Top characters: _esiona- > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Oct 22, 2004, at 9:17 AM, Jeff Abrahamson wrote: > > But I'm curious why this should be so. It's usually possible to reach > > a decision with more confidence if one has less data. More data adds > > nuance to decisions. Why should Bayesian filters (or Markovian or...) > > work worse if there's more data? > > Your second sentence here explains it quite well ;-) Basically, it > creates a higher signal-to-noise ratio w/r/t the tokens in your corpus. > A medium level of solid tokens is much better than a huge array of > who-knows-what tokens. I guess I meant that it's easier to make decisions without full information, but that doesn't necessarily mean the decisions are better. Just easier to make. -- Jeff Jeff Abrahamson <http://www.purple.com/jeff/> +1 215/837-2287 GPG fingerprint: 1A1A BA95 D082 A558 A276 63C6 16BF 8C4C 0D1D AE4B A cool book of games, highly worth checking out: http://www.amazon.com/exec/obidos/ASIN/1931686963/purple-20 Attachment:
signature.asc
|
|