[Corpora-List] Precision and Recall

[Angus Grieve-Smith]

On Tue, 22 Apr 2008, J Washtell wrote:

> I doubt most people look much beyond the first few pages of two Google, 
> whether the results are good or not, so it would seem reasonable to 
> frame ranking as filtering and call the first few pages the result set. 
> Is it *really* a special case?

 	I don't think it's a special case at all.  I disagree that 
filtering is a very good model for what people are doing, though, because 
I know *I* occasionally go through all the pages of Google results, 
sometimes dozens of pages.  It brings in the "long tail" of people who are 
so desperate to find more data that they'll spend the time to sift through 
lots of false positives - instead of deciding for them how much they care.


> Isn't it likely that this was more a problem with whatever proxy you 
> were using for customer satisfaction, than precision and recall per se? 
> If, for example, you were measuring "number of customers satisfied" 
> against "number of customers served", you ought to find that these sorts 
> of metrics do their job quite well... I'd have thought.

 	What I was suggesting was that management may have been using 
precision and recall as their measures of customer satisfaction; you can 
imagine the problems that that would cause.  But I shouldn't speculate, 
since I was a low-level worker bee and didn't have much access to what 
management was doing.

 					-Angus B. Grieve-Smith
 					grvsmth at panix.com

_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora