Hi Doug, comments below: (01)
-Rich (02)
Sincerely,
Rich Cooper
EnglishLogicKernel.com
Rich AT EnglishLogicKernel DOT com
9 4 9 \ 5 2 5 - 5 7 1 2 (03)
-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of doug foxvog
Sent: Monday, March 14, 2011 10:25 PM
To: ontolog-forum
Subject: Re: [ontolog-forum] Using controlled natural languages for ontology (04)
On Mon, March 14, 2011 12:17, Rich Cooper said:
> DF: "Events and situations would be semantic primitives."
>
> RC: It may be that specific TYPES of events and situations are primitive,
> but events and situations, at the full expression of those symbols in
> language, are not SEMANTIC - they are syntactic just as surely as verbs
> and nouns are syntactic. (05)
DF: I hold that the concepts of Event and Situation are semantic concepts
since relations can be defined for them that apply to their subtypes.
An Event would have participants, specialized into doer, objectActedOn,
output, and other such relations. Rules can be written for such relations.
Both Event and Situation would be subclasses of TemporalThing, allowing
relations described for such more-general concepts (such as startingTime,
duration, and endingTime) to (transitively) be applicable. (06)
DF: An upper ontology can certainly specialize the semantic categories of
Event and Situation, for example including SpatialEvent, MentalEvent,
and IntentionalAction. (07)
DF: Syntax, on the other hand depends on language. Many languages have both
nouns and verbs, while others, such as some native American languages,
do not. The same idea can be expressed using different parts of speech
for the same intrinsic meanings. (08)
"The War of 1812 lasted from 1812 to 1815."
"The duration of the War of 1812 was from 1812 to 1815."
"The US & UK fought from 1812 to 1815." (09)
DF: Syntactically, "lasted" in the first sentence is a verb, while
"duration" in the second sentence is a noun, but they play
the same semantic role. The verb "fought" in the third sentence
refers to the same event as the common noun phrase in the first
two sentences even though different parts of speech are used. (010)
> It isn’t the event or situation in GENERAL that expresses semantic
> values. Semantics is not GENERAL; it is very specific to
> the person interpreting events and situations. (011)
DF: An upper ontology describes GENERAL semantic classes and relations.
Mid-level and lower ontologies describe more specific types of things
and even individuals. Knowledge bases using such ontologies express
very specific semantics. (012)
-- doug foxvog (013)
RC: Thanks - that was a very well expressed formulation. Do you have a top
level ontology which you think is generally useful, and which you would
recommend WITHOUT having to cart along a lot of philosophical baggage and
historical impedimenta stuff? I just want a useful, fairly general ontology
for English sentences. I don't care if Ghengis Khan and Christopher
Columbus collaborated through a Oujia board to create philosophy. (014)
RC: Reading "The Stuff of Thought", by Steven Pinker, shows the linguistics
side of how American English speakers appear to construe meaning. He also
seems to use 4D conceptualizations in that book VERY occasionally, but does
it from a very practical viewpoint, the kind of viewpoint that can help
developers create realistic English functionality for various applications. (015)
RC: For example, patent claim language is very structured, yet with
occasional grammatical errors even so, and with nearly four decades of
patents online, they make good grist for analyzing event and situation
structures. Furthermore, each claim has specific key words, which can be
matched against the textual description explaining how the patent
accomplishes its intended end. (016)
RC: Looking at 49 patents in class 707, I find that NEARLY each one has a
unique set of keywords, but they cluster - maybe 15% of them have exactly
the same set of key words as one or more other patents. But the semantics
of interpreting the claim sentence has to be demonstrated in the patent, at
least to the point where a "person of ordinary skill in the art" could,
"without undue experimentation" practice the invention. (017)
RC: So the events and the situations described in patent claims are
expressed in English that has been thoroughly refereed, yet still has
occasional human errors. Each claim also has an ontology, about one per
claim, that is expressed in a common English sentence. In that form, a
patent claim ties events and situations together with variables and
constants of designation that forms an expressive sentence which is unique
within the patent claim collection. (018)
RC: The point is that there is NO upper ontology there. The patent
database is comprised of hundreds of thousands of documents. Yet over the
200 plus years, the culture in the USPTO has developed uniform ways of
classifying, organizing, comparing and deeply examining the novelty and
value of ONLY individual patents. (019)
RC: Yet events and situations abound in every patent, for every claim,
without exception. So my view is that the casting of events and situations
into top-of-the-lattice, abstracter events and abstracter situations than
have actually been observed, is not a proper rendering of reality. (020)
RC: I can see that, as scientists, it is useful to generalize, as well as
to specialize. But what concerns me is the JUSTIFICATION that said
generalization requires, but which I never see offered. Why in the world
would I think a generalization is REAL if I can't see it, can't sense it,
can't even build a sensor to test it? Generalization is a hypothesized
method of forming a new object from an old one, which is often very, very
useful in practice, so it is used. Even though we don't know what Pi is to
the very last digit, we also don't have such a justification for
generalizing all the way up to ONE TOP LEVEL. (021)
RC: It is more likely that there are numerous REAL, actually observed
events and situations that are observed, and which properly fit the
definition of a generalization of other events and situations that have also
been confirmed by observation in reality. But to run it to the limit, and
claim that any generalization any time any where is justified, as well as
simply is logical in the extreme, is not to my liking. There needs to be a
rationale about when to stop, or the whole process cannot be called logical
in the full sense of my use of that word. (022)
JMHO,
-Rich (023)
> -Rich
>
> Sincerely,
> Rich Cooper
> EnglishLogicKernel.com
> Rich AT EnglishLogicKernel DOT com
> 9 4 9 \ 5 2 5 - 5 7 1 2
>
> -----Original Message-----
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
> [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of doug foxvog
> Sent: Sunday, March 13, 2011 8:36 PM
> To: [ontolog-forum]
> Subject: Re: [ontolog-forum] Fwd: Re: Using controlled natural languages
> for
> ontology
>
> On Sat, March 12, 2011 15:50, FERENC KOVACS said:
>> JS: That ambiguity is not "inherent" in NLs,
>> but it is the result of using NLs by an open-ended community of
>> people with different backgrounds, needs, interests, and goals.
>>
>> FK This fact is represented in connections called relations - other tah
>> Boolean, or algeraic.
>> So the real problem is to find a common denominator for making domain
>> ontologies
>> interoperable. The job has to be started from top down, provided that
>> the
>> conceptualisation is correct. Looking at SUMO and the Italian profesors
>> slides I maintain that the model is bleeding.
>
>
>> JS: I do not believe that a broad-coverage detailed ontology of
>> everything
>> is possible or desirable.
>
>> FK You need not have to have a broad-coverage detailed ontolopgy of
>> eevrything -
>
> You both agree.
>
>> if you do not stick to the way your semantic primitives, i.e the concept
>> of objects, properties and relations are used to make up a model with
>> non semantic primitives (including events, situations etc.!!).
>
> Events and situations would be semantic primitives.
>
>> Among other things it is not clear for a great many people
>> that there are three different variables that all have independent lives
>> to be synchronised and regularly updated, namely the objects themselves,
>
> You mean the objects within the ontology, here, right?
>
>> the names of such objects (the sign, token, denotator paradgims)
>
> I note that these names are names of parts of the ontology, not
> names of the things they refer to. Such names could be simple
> within the ontology, but when referenced from elsewhere would
> appropriately be URIs. In many languages, they could be referenced
> by a namespace identifier followed by a name local to the ontology.
>
> Note that such names need not be similar to natural language words.
> snwmd:362183003 would be an acceptable name for the entire left lobe
> of the liver.
>
> An ontology should have mappings to one or more text strings which
> may represent the intended meaning of the thing represented by each
> object in the ontology as well as a comment that clearly describes
> that intended meaning.
>
>> and teh concepts of objects that are in the mind of the poeple
>
> Certainly, the concepts that are in peoples minds are different from
> the objects in the ontologies, and the names of those objects. All
> of these are different from both the objects (or types) that those
> concepts refer to and the names used in various languages and jargons
> to refer to the referents.
>
>> -
>> naturally all different. So what you have in addition to those three is
>> the name of the concepts and the properties of the objects that are
>> referred to by that name (sign) and by that concept.
>
> Sure. It is important to distinguish these things:
> * Object in ontology
> * Name of object in ontology
> + local name
> + global name (including name of ontology)
> * Thing meant by the ontology object
> * NL names of thing meant by ontology object
> * Concept in people's minds of the thing meant by ontology object
> + Often there would be no need to model this in an ontology
>
> Note that properties of things modeled in ontologies should themselves
> be modeled by objects in the ontology. These ontology objects would
> be predicates in the ontology as opposed to the class and individual
> objects. These predicate objects would be named within the ontology
> and should be mapped to NL words or phrases which indicate the
> relation.
>
>> Unless you do not have a system or model to align
>> your different concpets in your different frame of mind through a
>> common notation for the same objects (identity) through
>> the use of the same wording and the same phsical object reference
>> (timespace) also tied together, you suck
>
> I think you mean "you're stuck", here.
>
> Mappings would be needed among the ontology objects, referred to by
> their names (within the ontology) and NL names for the objects. I'm
> not sure what else you are referring to here. Are you referring to
> mapping between terms in different ontologies? Are you referring to
> establishing a mapping between an ontology term and an object in the
> physical world (instead of merely to its referent in the ontology)?
>
> -- doug
>
>> Regards,
>> Ferenc
>> "our attention becomes structured by external demands. In more intimate
>> encounters, the level of both challenges and skills can grow very high.
>> Thus,
>> interactions have many of the characteristics of flow activities, and
>> they
>> certainly require the orderly investment of mental energy. The strong
>> effects of
>>
>> companionship on the quality of experience suggest that investing energy
>> in
>> relationships is a good way to improve life." Mihály
>>
>
Csíkszentmihályi"http://www.psychologytoday.com/articles/199707/finding-flow
> ?page=3
>>
>>
>> _________________________________________________________________
>> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
>> Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
>> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
>> Shared Files: http://ontolog.cim3.net/file/
>> Community Wiki: http://ontolog.cim3.net/wiki/
>> To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
>> To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>>
>>
>
>
> =============================================================
> doug foxvog doug@xxxxxxxxxx http://ProgressiveAustin.org
>
> "I speak as an American to the leaders of my own nation. The great
> initiative in this war is ours. The initiative to stop it must be ours."
> - Dr. Martin Luther King Jr.
> =============================================================
>
>
> _________________________________________________________________
> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
> Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
> Shared Files: http://ontolog.cim3.net/file/
> Community Wiki: http://ontolog.cim3.net/wiki/
> To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
> To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>
>
> (024)
=============================================================
doug foxvog doug@xxxxxxxxxx http://ProgressiveAustin.org (025)
"I speak as an American to the leaders of my own nation. The great
initiative in this war is ours. The initiative to stop it must be ours."
- Dr. Martin Luther King Jr.
============================================================= (026)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (027)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (028)
|