TOP and SCORE

[Arnold Rots]

My point is that input and output are dealing with the same quantity
and usually that means that we use the same parameter name in order
not to obfuscate things.
More specifically, TOP refers to the relatve values of SCORE and is
thus a function of SCORE.
What we have implemented for input is, as I said:
  SCORE=ALL
  SCORE=TOP
It is clear from this that "TOP" refers to the SCORE which is not the
case for a bare TOP parameter.
One might easily extend these values to:
  SCORE=TOP5
and:
  SCORE=BOTTOM3

When all this was first proposed, there was discussion of whether the
parameter name should be GRADE, SCORE, or RANK.  We settled on SCORE
and we changed our implementation accordingly.  Don't keep changing
it, please.

Cheers,

  - Arnold

Markus Dolensky wrote:
> My understanding is that SCORE is a measure of the relevance of a 
> matching record relative to the other records in the result set (and 
> there has been some discussion about calling it RANKING instead).
> 
> An example:
> 
> input:  TOP=3
> output: SCORE={7.5, 76.3, 0.003}
> 
> In this example SCORE is expressing the fact that the expected relevance 
> of the records with respect to the query parameters is very different. 
> Hence, in this context it is certainly useful.
> 
> Patrick Dowler wrote:
>  > PS-To be complete, we could also have BOTTOM, because sometimes you
>  > want to get the worst results :-)
> 
> In order to implement Pat's suggestion I propose to use negative 
> integers for expressing BOTTOM, e.g., ...
> 
> TOP=-3
> 
> ... to list the three least relevant records.
> 
> Greetings,
> Markus
> 
--------------------------------------------------------------------------
Arnold H. Rots                                Chandra X-ray Science Center
Smithsonian Astrophysical Observatory                tel:  +1 617 496 7701
60 Garden Street, MS 67                              fax:  +1 617 495 7356
Cambridge, MA 02138                             arots at head.cfa.harvard.edu
USA                                     http://hea-www.harvard.edu/~arots/
--------------------------------------------------------------------------