W3C home > Mailing lists > Public > semantic-web@w3.org > March 2008

Re: semantic definition

From: Golda Velez <gv@btucson.com>
Date: Mon, 10 Mar 2008 15:03:08 -0700
To: "Azamat" <abdoul@cytanet.com.cy>
Cc: "'SW-forum'" <semantic-web@w3.org>, "[ontolog-forum] " <ontolog-forum@ontolog.cim3.net>
Message-Id: <200803101503.09871.gv@btucson.com>

On Saturday 08 March 2008 12:34, Azamat wrote:
> I was surprised that the issues of definition, its nature, kinds, meaning, 
> demonstration, and formalization, has not received a separate thread and 
> careful discussion both on the semantic web and ontology forums. Since it is 
> hardly to find another notion so decisive for ontology and semantic web 
> ...and so muddled and vague.

Thanks, Azamat!  This post really got me thinking, though in a bit of a 
different direction.   I'm not sure what we need is more formal definition, 
though in some areas that would certainly help, but more a more flexible type 
of inference model that allows for conflicting information and partial 
matches.

I can see a few people using Neural Nets to develop ontologies, but is anyone 
out there using or interested in using neural net type models to generate 
sort of stable-state inferences?   I'm probably not being very clear, but 
with a model of weighting for similarities of initial subject and weighting 
the stability of the 'inferred state' from the initial state maybe you could 
approach the whole complex logic thing in a different way?   I can see that 
some people are already working on fuzzy assertions, did that get accepted? 

I found this

Extending OWL by fuzzy description logic
http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?tp=&arnumber=1562993&isnumber=33171

and this

A Fuzzy Extension of SWRL
http://www.w3.org/2004/12/rules-ws/paper/52/

At least Stamou et all seem to agree with me ;-)  : "The need to deal with 
imperfect and imprecise information is likely to be common in the context of 
the (Semantic) Web."  I'll see if I can find out if anything came of that...

I took some classes from Hopfield (mr neural net) way back when, and I've 
always thought that neural nets are a reasonable model of how people think.  
If we're trying to model human knowledge, it makes sense to model it in a way 
that human brains may work...we're capable of precise logic, but I'm not sure 
all our best work is done only that way.

--Golda

> For currently the definitions [as giving the  
> meanings of expressions, symbols, constructs, or things] might be 
> constructed by reference to:
> 
> classification or dichotomy (classificatory definition);
> 
> properties (genus&differentia) (essential definition);
> 
> parts, accidental properties (ostensive definition)
> 
> context (contextual definition);
> 
> cause, genesis, origin (genetic or genealogical definition);
> 
> end or purpose (stipulative definition);
> 
> interest (pragmatic definition);
> 
> common use (dictionary definition);
> 
> induction (recursive definition);
> 
> intension (connotation) (intensional or connotative definition);
> 
> extension (denotation) (extensional definition).
> 
> 
> Such inconsistency of definition is a cause of diverse views of meaning 
> itself, which is defined as in:
> 
> idea, thought or intention;
> 
> operation, measurement, and computation;
> 
> usage, utility;
> 
> truth condition;
> 
> sense, intension, connotation, content;
> 
> reference, extension, denotation;
> 
> sense and reference together.
> 
> A definition may be qualified as consistent and contradictory, true or 
> false, arbitrary or real, proved or unproved, accidental or essential, 
> formal or material, nominal or real.
> 
> 
> It is stated that many axioms of the sciences, formal and theoretical, are 
> nothing but definitions in disguise. That definition is formal and precise 
> unlike description, explanation, interpretation. And that formally it is a 
> kind of an equivalence relation where the left side (the definiendum, that 
> which is to be defined) is a function of the right side (definiens, that 
> which defines, determines, specifies).
> 
> 
> 
> The most prospective method of defining seems to be a semantic real 
> definition, where the definition involves the primary meaning (key 
> denotation cum major connotation) of a word, a phrase, a symbol, a concept 
> or an entity, so that it states the nature of the object defined and is 
> convertible with its subject. The semantic definition is a relatively stable 
> construct. Although, it is liable to the controlled redefinition as far as 
> the knowledge of the world progresses, but this updating should not be 
> something unpredicted or unforeseen, a radically new definition.
> 
> 
> 
> There is an example of motherhood, the family relationship between an 
> offspring and the female parent. Presently, a child may have mothers as 
> diverse as natural mother, biological mother, adopting mother, step mother, 
> surrogate mother, cloning mother, etc. When we use a standard nominal 
> definition of mother as {''a woman giving birth to a child''), we are 
> missing all possible kinds of motherhood. When we use a real definition (''a 
> woman parenting (producing, begetting or raising) a child''), verified by 
> experimentation, we can cover all the key senses of motherhood.
> 
> 
> 
> Re ''the personal representations of reality'', being a sort of nominal 
> definitions and usually having nothing to do with the nature of things, such 
> definitions are not harmful as far as they used for specific personal 
> purposes, but most harmful then presented as the scientific and objective 
> definitions. One may define erotica as '' a creative activity to stimulate a 
> reproductive activity'', capitalism as '' an economic system based on 
> exploitation and profiteering'', political party as ''an organization to 
> gain power by revolution'', business as ''a commercial activity to 
> profiteer'', etc., thus creating an unreal world.
> 
> The same reasoning applies to defining ontology, ontological classes, 
> computing ontology, ontology engineering, ontology languages and tools, 
> semantics, semantic concepts, semantic systems, semantic web, semantic web 
> technologies, etc.
> 
> Here I met a conference definition of semantic interoperability as β€˜β€™the 
> common automatic interpretation of the meaning of the exchanged information, 
> i.e. the ability to automatically process the information in a 
> machine-understandable manner.’’ Now I wonder what sort of definition it 
> might be: nominal, extensional, pragmatic, ostensive.
> 
> 
> 
> azamat abdoullaev
> 
> 
> 
> 
> 
>  
> 
> 

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Golda Velez	520-440-1420		http://goldavelez.com
what I do: 	Tucson Superblog	http://btucson.com
		Search software		http://webglimpse.net
		Web hosting		http://iwhome.com

"Help organize the world - index your own corner of the web!"
Received on Monday, 10 March 2008 21:52:58 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:42:03 UTC