RE: [UCR] RIF needs different reasoning methods

[just a quick thanks]
>> > Isn't resolution a refutation technique? Did you mean 
>> > something stronger?
>>
>> REsolution (including SL resolution) are refuytartion methods. SLD
>> resolution is not - in spite of a common belief.
>
> Francois is pointing to the fact that SLD resolution corresponds 
> "isomorphically" to constructive inference based on Modus Ponens.
> Therefore, it's not really refutation-based.

Thanks Gerd for pointing this out; I still learn so much
and was kind of just doing this for many years without
realizing; indeed the kind of proof structures that I get
from instrumenting an SLD type of reasoning is just GMP
i.e. in a Reverse Polish Notation style compact proof (*)
"push the facts, pop the antecedents and push the conclusions". 


-- 
Jos De Roo, AGFA http://www.agfa.com/w3c/jdroo/

(*) e.g. the "body" tuples of the objects with "head":"GND"
in http://eulersharp.sourceforge.net/2006/02swap/graph.html

Received on Monday, 6 March 2006 21:41:07 UTC