Predictive or scoring solution for PostgreSQL ?

Hi,

Does anyone know a predictive or a database scoring solution for PostgreSQL ?

I'm looking for a system able to take a database with for example 100 000
records in total, inside them we have got 1000 records with one field set to
YES ... with about 100 fields in the table ...
The system should be able to set a score to the 100 fields to determine the
most importants fields to this 1000 records who's got the YES value ...
Then set a formula ... to calculate and to apply to the rest of the database
the same score ... and then estimate (predictive thing) in the 90 000 rest of
records which one may have the famous field set to YES ...

I hope I'm clear in my demand ... ;o)

Hope also someone have already heard about this ... and may be could help
me ;o)

best regards,
--
Hervé