GiST indices and statistical approximations

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi, I was reading the Wikipedia entry on GiST (http://
en.wikipedia.org/wiki/GiST) and it mentions

"""
Although originally designed for answering Boolean selection queries,
GiST can also support nearest-neighbor search, and various forms of
statistical approximation over large data sets.
"""

Could anybody provide pointers to papers or online pages that discuss
the use of GiST indexes for statistical approximations?

Thanks,

- -------------------------------------------------------------------
Rajarshi Guha <rguha(at)indiana(dot)edu>
GPG Fingerprint: D070 5427 CC5B 7938 929C DD13 66A1 922C 51E7 9E84
- -------------------------------------------------------------------
A memorandum is written not to inform the reader,
but to protect the writer.
-- Dean Acheson

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (Darwin)

iEYEARECAAYFAkiZF38ACgkQZqGSLFHnnoQ36QCfeaU6mDid4v5DDH4V2vDMdaZr
At8AoJQj9ZHxDbGiarfwNHe9aIYr1pnB
=Jrgz
-----END PGP SIGNATURE-----