Text Clustering Software working with Protege OWL

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Text Clustering Software working with Protege OWL

Peter-336
I am working on processing the papers (both physical and electronic) of
Professor Richard Fikes in Stanford University. I would like to use text
clustering software to process the emails because there are thousands of them.
Questions:

1 I am wondering if I create an ontology on the domain of Professor Fikes' paper
(include authors, co-authors, aboutness (topics, projects, technology) of the
writings, etc.), will I get a better result from the text clustering software.

2. Is there a open source text clustering software which work with Protege OWL?

Any help will be very much appreciated.

Thanks & Regards,

Peter

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 
Reply | Threaded
Open this post in threaded view
|

Re: Text Clustering Software working with Protege OWL

adam.saltiel
Peter,
What do you mean by text clustering software, what do you have in mind?
Are there any clustering solutions that work directly with hand crafted ontologies?
I would have thought that clustering software - if you mean some sort of SVD or LSI solution - would present terms for an ontology, the issue would be the relationship between significant terms and phrases as discovered by, say, LSI, and an ontology that makes sense of the knowledge domain as the two may not be coincident.
What do you think?
Adam

 
On 10/04/2008, Peter <[hidden email]> wrote:
I am working on processing the papers (both physical and electronic) of
Professor Richard Fikes in Stanford University. I would like to use text
clustering software to process the emails because there are thousands of them.
Questions:

1 I am wondering if I create an ontology on the domain of Professor Fikes' paper
(include authors, co-authors, aboutness (topics, projects, technology) of the
writings, etc.), will I get a better result from the text clustering software.

2. Is there a open source text clustering software which work with Protege OWL?

Any help will be very much appreciated.

Thanks & Regards,

Peter

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 
Reply | Threaded
Open this post in threaded view
|

Re: Text Clustering Software working with Protege OWL

Peter-336
In reply to this post by Peter-336

Hi Adam,
 
Sorry for the ignorance on clustering software. I am not familiar with
either SVD or LSI. What I have in mind is to handcraft an ontology using
Protege OWL for my special need and use "text clustering" software to
categorize (most important requirement) and also to assign properties
(secondary important) to documents (emails, papers, etc.) using my
ontology in addition to the what are already included in the software.
Ideally, the "text clustering" software can also suggest terms
for including to the ontology with my approval.
I hope this will clarify my requirements.
 
Thanks
>

> Message: 3
> Date: Thu, 10 Apr 2008 22:28:52 +0100
> From: adasal <[hidden email]>
> Subject: Re: [protege-owl] Text Clustering Software working with
> Protege OWL
> To: "User support for the Protege-OWL editor"
> <[hidden email]>
> Message-ID:
> <[hidden email]>
> Content-Type: text/plain; charset="utf-8"
>
> Peter,
> What do you mean by text clustering software, what do you have in mind?
> Are there any clustering solutions that work directly with hand crafted
> ontologies?
> I would have thought that clustering software - if you mean some sort of SVD
> or LSI solution - would present terms for an ontology, the issue would be
> the relationship between significant terms and phrases as discovered by,
> say, LSI, and an ontology that makes sense of the knowledge domain as the
> two may not be coincident.
> What do you think?
> Adam
>
>
> On 10/04/2008, Peter <[hidden email]> wrote:
> >
> > I am working on processing the papers (both physical and electronic) of
> > Professor Richard Fikes in Stanford University. I would like to use text
> > clustering software to process the emails because there are thousands of
> > them.
> > Questions:
> >
> > 1 I am wondering if I create an ontology on the domain of Professor Fikes'
> > paper
> > (include authors, co-authors, aboutness (topics, projects, technology) of
> > the
> > writings, etc.), will I get a better result from the text clustering
> > software.
> >
> > 2. Is there a open source text clustering software which work with Protege
> > OWL?
> >
> > Any help will be very much appreciated.
> >
> > Thanks & Regards,
> >
> > Peter
> >
> > _______________________________________________
> > protege-owl mailing list
> > [hidden email]
> > https://mailman.stanford.edu/mailman/listinfo/protege-owl
> >
> > Instructions for unsubscribing:
> > http://protege.stanford.edu/doc/faq.html#01a.03
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: http://mailman.stanford.edu/pipermail/protege-owl/attachments/20080410/b80a7748/attachment.htm
>
> ------------------------------
>
> _______________________________________________
> protege-owl mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-owl
>
>
> End of protege-owl Digest, Vol 21, Issue 34
> *******************************************



Going green? See the top 12 foods to eat organic.
_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 
Reply | Threaded
Open this post in threaded view
|

Re: Text Clustering Software working with Protege OWL

adam.saltiel
AFAIK you will have to get down and dirty with the code. You might want to start with Lucene.
The problem is that to cluster you must inspect and to inspect there must be some method or basis. Lucene will index, then you can retrieve from the index and cluster using something like SVD, for which there is an implementation.
Then you can judge whether what is coming up bears any resemblance to what you would expect from the logical perspective of an Ontology.
I expect that terms that are close (proximate) to your key terms, the Ontology classes, would be candidates as properties of the class?
It might be helpful to cluster around phrases rather than single terms.
I would be interested if you find anything that does this out of the box that is open source. And I doubt there is a commercial solution that works with Protege as you asked.
You might like to read the "LSI, Latent Semantic Indexing" thread from 29/01/2007 on the [hidden email] list.
I will follow this thread to see what luck you have.
Adam

On 12/04/2008, chan peter <[hidden email]> wrote:

Hi Adam,
 
Sorry for the ignorance on clustering software. I am not familiar with
either SVD or LSI. What I have in mind is to handcraft an ontology using
Protege OWL for my special need and use "text clustering" software to
categorize (most important requirement) and also to assign properties
(secondary important) to documents (emails, papers, etc.) using my
ontology in addition to the what are already included in the software.
Ideally, the "text clustering" software can also suggest terms
for including to the ontology with my approval.
I hope this will clarify my requirements.
 
Thanks
>
> Message: 3
> Date: Thu, 10 Apr 2008 22:28:52 +0100
> From: adasal <[hidden email]>
> Subject: Re: [protege-owl] Text Clustering Software working with
> Protege OWL
> To: "User support for the Protege-OWL editor"
> <[hidden email]>
> Message-ID:
> <[hidden email]>
> Content-Type: text/plain; charset="utf-8"

>
> Peter,
> What do you mean by text clustering software, what do you have in mind?
> Are there any clustering solutions that work directly with hand crafted
> ontologies?
> I would have thought that clustering software - if you mean some sort of SVD
> or LSI solution - would present terms for an ontology, the issue would be
> the relationship between significant terms and phrases as discovered by,
> say, LSI, and an ontology that makes sense of the knowledge domain as the
> two may not be coincident.
> What do you think?
> Adam
>
>
> On 10/04/2008, Peter <[hidden email]> wrote:
> >
> > I am working on processing the papers (both physical and electronic) of
> > Professor Richard Fikes in Stanford University. I would like to use text
> > clustering software to process the emails because there are thousands of
> > them.
> > Questions:
> >
> > 1 I am wondering if I create an ontology on the domain of Professor Fikes'
> > paper
> > (include authors, co-authors, aboutness (topics, projects, technology) of
> > the
> > writings, etc.), will I get a better result from the text clustering
> > software.
> >
> > 2. Is there a open source text clustering software which work with Protege
> > OWL?
> >
> > Any help will be very much appreciated.
> >
> > Thanks & Regards,
> >
> > Peter
> >
> > _______________________________________________
> > protege-owl mailing list
> > [hidden email]
> > <a href="https://mailman.stanford.edu/mailman/listinfo/protege-owl" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">https://mailman.stanford.edu/mailman/listinfo/protege-owl
> >
> > Instructions for unsubscribing:
> > <a href="http://protege.stanford.edu/doc/faq.html#01a.03" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">http://protege.stanford.edu/doc/faq.html#01a.03
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <a href="http://mailman.stanford.edu/pipermail/protege-owl/attachments/20080410/b80a7748/attachment.htm" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">http://mailman.stanford.edu/pipermail/protege-owl/attachments/20080410/b80a7748/attachment.htm
>
> ------------------------------
>
> _______________________________________________
> protege-owl mailing list
> [hidden email]
> <a href="https://mailman.stanford.edu/mailman/listinfo/protege-owl" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">https://mailman.stanford.edu/mailman/listinfo/protege-owl
>
>
> End of protege-owl Digest, Vol 21, Issue 34
> *******************************************



Going green? <a href="http://green.msn.com/galleries/photos/photos.aspx?gid=164&amp;ocid=T003MSN51N1653A" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">See the top 12 foods to eat organic.

_______________________________________________
protege-owl mailing list
[hidden email]
<a onclick="return top.js.OpenExtLink(window,event,this)" href="https://mailman.stanford.edu/mailman/listinfo/protege-owl" target="_blank">https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: <a onclick="return top.js.OpenExtLink(window,event,this)" href="http://protege.stanford.edu/doc/faq.html#01a.03" target="_blank">http://protege.stanford.edu/doc/faq.html#01a.03



_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03