ontolog-forum
[Top] [All Lists]

Re: [ontolog-forum] Ontology extraction tool

To: "David Decraene" <David@xxxxxxxxxxxxxxx>
Cc: "[ontolog-forum] " <ontolog-forum@xxxxxxxxxxxxxxxx>, Mariana Casella Dos Santos <mariana@xxxxxxxxxxxxxxx>
From: Pat Hayes <phayes@xxxxxxx>
Date: Wed, 4 Oct 2006 12:26:51 -0500
Message-id: <p06230903c1499cf387d3@[10.100.0.128]>
>I completely disagree with this.
>L&C's technology possesses NLP capabilities. In 
>specific, we have automated analysis of free 
>text, it translates the terms found into 
>existing (or non-existing) ontology concepts and 
>is able introduce meaningful relations.    (01)

I am sure that your tool can extract some 
content, with some degree of accuracy, from some 
kinds of texts. I am also sure that it cannot 
extract all the content with full accuracy from 
arbitrary free text. Exactly what it is capable 
of can only be determined by experiment, 
preferably conducted in an open peer-reviewed 
scientific setting. Do you have any empirical 
results to report?    (02)

>Our NLP engine (Tessi) detects paragraphs, finds 
>sentence boundaries, uses a dependency parser to 
>get a correct parse-tree, and uses a semantic 
>tagger to infer ontology concepts from these 
>parsetrees. The ontocreation module is able to 
>construct new ontology content based on what was 
>parsed as well.
>
>Our lexicon (the language layer connecting text 
>to ontology) is integrated into our very own 
>formal ontology, named linkbase (700.000 core 
>concepts).    (03)

What kinds of inference can your formal ontology 
support? What logic is it written in?    (04)

>It's not an AI miracle, it's a formal ontology miracle.    (05)

If it did what you seem to be claiming it can do, 
it would be an AI miracle. I however do not 
believe in miracles of any stripe. The history of 
AI is littered with somewhat overinflated claims. 
While recent work has produced impressive gains 
in performance, the world is still a long way 
from having a fully competent free text 
comprehension system. I realize that commercial 
pressure may require you to focus on the positive 
side of things, but IMO we should all guard 
against overinflated claims, which tend to 
produce backlash.    (06)

BTW, I would be delighted to be proved wrong; but 
I have high standards of proof.    (07)

Pat Hayes    (08)

>
>David Decraene
>PH.D. BioMedical sciences/Ontology modeller.
>Language and Computing NV
>Kortrijksesteenweg 1038
>9051 Sint-Denijs-Westrem
>
>
>-----Original Message-----
>From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
>[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx]On Behalf Of Pat Hayes
>Sent: dinsdag 3 oktober 2006 23:25
>To: Arsic, Antoinette
>Cc: [ontolog-forum]
>Subject: Re: [ontolog-forum] Ontology extraction tool
>
>
>>A fellow MITRErite was asking the following, and I thought this group
>>was the best you can get to ask:
>>
>>(I'm) looking for an automated ontology extraction tool that can be
>>used against unstructured data
>
>What exactly do you mean by 'using against' here?
>If you want a tool that can input unstructured
>text input and output a formal structured
>ontology representing the same information, then
>no such tool exists or will exist in the
>forseeable future. That is asking for an AI
>miracle. If anyone offers you one, ask for the
>Brooklyn Bridge instead.
>
>Pat Hayes
>
>>(softcopy text documents in MS Word,
>>EXCEL or PDF formats) and structured data.  Is anyone aware of such a
>>tool(s)?
>>
>>
>>------------------------------
>>Antoinette Arsic
>>The MITRE Corporation
>>703-983-5286 (office)
>>443-567-2703 (cell)
>>
>>
>>
>>
>>
>>_________________________________________________________________
>>Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/ 
>>Subscribe/Unsubscribe/Config:
>>http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/ 
>>Shared Files: http://ontolog.cim3.net/file/
>>Community Wiki: http://ontolog.cim3.net/wiki/
>>To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>>
>
>
>--
>---------------------------------------------------------------------
>IHMC           (850)434 8903 or (650)494 3973   home
>40 South Alcaniz St.   (850)202 4416   office
>Pensacola                      (850)202 4440   fax
>FL 32502                       (850)291 0667    cell
>phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
>
>
>_________________________________________________________________
>Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/ 
>Subscribe/Unsubscribe/Config: 
>http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/ 
>Shared Files: http://ontolog.cim3.net/file/
>Community Wiki: http://ontolog.cim3.net/wiki/
>To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>
>
>_________________________________________________________________
>Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/ 
>Subscribe/Unsubscribe/Config: 
>http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/ 
>Shared Files: http://ontolog.cim3.net/file/
>Community Wiki: http://ontolog.cim3.net/wiki/
>To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>    (09)


-- 
---------------------------------------------------------------------
IHMC            (850)434 8903 or (650)494 3973   home
40 South Alcaniz St.    (850)202 4416   office
Pensacola                       (850)202 4440   fax
FL 32502                        (850)291 0667    cell
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes    (010)


_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Subscribe/Unsubscribe/Config: 
http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (011)

<Prev in Thread] Current Thread [Next in Thread>