Andreas,
> Relevant fields could be name, street, zip, city, phone
> Is there a way to do something like this with postgresql ?
> I fear this will need still a lot of manual sorting and searching even when
> potential peers get automatically identified.
One of the techniques I use to increase the odds of detecting
duplicates is to trim each column, remove all internal whitespace,
coalesce it into a single string, and calculate an MD5 (some other
hash function may be better) hash. It's not perfect (we are dealing
with humans, after all), but it helps.
-- Gary Chambers
/* Nothing fancy and nothing Microsoft! */