Wednesday, March 28, 2012
Need Help 2
I am currently trying to figure out how to code this
I need the count of all rows in colum TERM that is > 1 when its grouped by the COLUM NAME
Anyone got any Ideas,i'm not sure wheter i get it right.
hope that it was the thing that you ask for.
Select TERM,Count(<table key>) as Cnt from <Table \name>
Wher TERM > 1
Group by TERM.
If I didn't get it right then just forget about it.
:D|||select count(*) from tbl
where term in
(
select term from tbl group by name having count(*) > 1
)
Monday, February 20, 2012
Near Proximity Term Questions
1. How near is near? Measured in characters, words, or whatever?
2. How do you know? Is this documented? Can't find it anywhere.
3. Can it be adjusted? at design time? at runtime?
I have used a program called Sonar which has powerful proximity options that allow the user to specify proximity in terms of words at runtime. Would like to be able to do that but can't find much on "near" in the documentation other than it seems to relative, provides for left and right nearness, and allows for chaining of multiple search terms.
Hi,
1. Near is mesured in words (depends on wordbreaker).
2. Please refer to a previous post in MS newsgroups : http://groups.google.com/group/microsoft.public.sqlserver.fulltext/browse_frm/thread/d2a3d0646ee6e68d/61db754db51cc5af?lnk=st&q=Near+keyword+in+Full+Text+Search+SQL+Server+2005&rnum=1&hl=en#61db754db51cc5af
3. NEAR ranking formulas are "by design"
Jean-Pierre Riehl
|||Hi Jean-Pierre,
Thanks for the very helpful info and link. A follow-up question if you don’t mind.
Since the size of the “Near” proximity parameter seems to be determined by the wordbreaker used by Sql Server's Full Text Search Engine (with a default value of 50 words), can the wordbreaker be altered (with a sql statement or otherwise) to increase or decrease the proximity of “Near” (i.e., make it more or less than 50 words)?