Using MX 6.1...I have a question regarding scoring in verity.  First a little 
background:

According to Mac docs, STEM and MANY are modifiers that are the default 
behavior of the CFSEARCH command (if you do not specify any other modifiers).  
This is what allows us to get a 'score' result giving us a 'relevancy' ranking 
of the results.  The Mac docs state that STEM causes similarly stemmed words to 
come up in the results (e.g. if you search on 'instructs', your results will 
also include words like 'instructional', instruction', etc.).  

Currently, the score is based on the number of occurrences of the search 
criteria in a document.  The more times it occurs in a document, the higher the 
score.  However, since we are using a STEM modifier, it's not just the exact 
criteria that is used to determine the score, it's anything with the same STEM. 
 So, there exists the possibility, if not probability, that a document with 4 
occurrences of the word 'instructional' would have a higher score than a 
document with 2 occurrences of the word 'instructs'.  

This is part of my problem.  Our client wants the benefits of STEM (including 
multiple similar terms in the search results) but wants documents with the 
'actual' criteria scored HIGHER than documents just having STEMS of the 
criteria.

Is this possible, and if so, how would you do it?  My thinking is that the only 
thing we could do is take the results from CFSEARCH and manipulate them 
manually to change the scores, but that would be a huge hit on processing.  I 
really need to know if you can do this somehow with modifiers in Verity that we 
can pass through the CFSEARCH interface.

Thanks!

Dave

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~|
Create robust enterprise, web RIAs.
Upgrade & integrate Adobe Coldfusion MX7 with Flex 2
http://ad.doubleclick.net/clk;56760587;14748456;a?http://www.adobe.com/products/coldfusion/flex2/?sdid=LVNU

Archive: 
http://www.houseoffusion.com/groups/CF-Talk/message.cfm/messageid:265498
Subscription: http://www.houseoffusion.com/groups/CF-Talk/subscribe.cfm
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4

Reply via email to