W3C home > Mailing lists > Public > semantic-web@w3.org > August 2011

Fwd: fact checking for semantic reasoners

From: Paola Di Maio <paola.dimaio@gmail.com>
Date: Mon, 29 Aug 2011 10:47:24 +0100
Message-ID: <CAMXe=SpvB5Jm_b_oRQdSyOqjNmsmH5v9zyQjSTBXSpht0AfdTQ@mail.gmail.com>
To: semantic-web at W3C <semantic-web@w3c.org>
> >
> What do you mean by "fact checking"?  Do you mean proving one assertion
> (or set of assertions) based on another set of assertions?
>

well, rather a 'verified' assertion. (verified by some ontology, presume)
its very easy to get things wrong, a perfectly valid syllogism
can lead to the wrong conclusion if the fact being asserted is wrong
(the truth preserving thing?)

if fred is a bat,
and all bats are birds (er.. looks like that, doesnt it?..)
then fred is a bird  (but if bats are not birds this conclusion would be
wrong)


...also thinking of all the false statements in wikipedia for example
that would need to be checked... simply because an editor
approves it (or requests a deletion of a fact) does not mean is true (or
false)

P

p




>
>
>
> --
>
> David Booth, Ph.D.
> http://dbooth.org/
>
> Opinions expressed herein are those of the author and do not necessarily
> reflect those of his employer.
>
>
Received on Monday, 29 August 2011 09:47:52 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 5 July 2022 08:45:25 UTC