On Sat, 8 Mar 2008, Jan Urbaski wrote:
>> Unfortunately, selectivity estimation for query is much difficult than just
>> estimate frequency of individual word.
> Sure, given something like 'cats & dogs'::tsquery the frequency of 'cat' and
> 'dog' won't suffice. But at least it's a starting point and if we estimate
> that 80% of the documents have 'dog' and 70% have 'cat' then we can tell for
> sure that at least 50% have both and that's a lot better than 0.1% that's
> being returned now.
certainly yes and given that most popular queries are single word query
this would very helpful in most cases.
The reason I though about ts_stat() improvement is that we could use its
statistics for incomplete search feature people requested, when
AND query like ( a & b &c ) rewrites to a set of AND|OR queries depending
on the terms occurency.
Oleg Bartunov, Research Scientist, Head of AstroNet (www.astronet.ru),
Sternberg Astronomical Institute, Moscow University, Russia
Internet: oleg(at)sai(dot)msu(dot)su, http://www.sai.msu.su/~megera/
phone: +007(495)939-16-83, +007(495)939-23-83
In response to
pgsql-hackers by date
|Next:||From: Warren Turkal||Date: 2008-03-09 08:32:20|
|Subject: timestamp datatype cleanup|
|Previous:||From: Oleg Bartunov||Date: 2008-03-09 02:30:57|
|Subject: Re: Google Summer of Code 2008 |