Cluster Scores & similarity

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

Cluster Scores & similarity

waleedAzmy
what does the cluster score represent? and Is there a normalized version from the score? if not how can I normalize it?

Can I get what is called the label score? if yes, how? what does it represents?  


Another questions,
how can I set the Label_Similarity_Threshold LINGO parameter?

Thank you.

Hope to help;
Reply | Threaded
Open this post in threaded view
|

Re: Cluster Scores & similarity

Dawid Weiss-2
> what does the cluster score represent? and Is there a normalized version from
> the score? if not how can I normalize it?

Each algorithm has its own notion of a cluster's score, so it does not
have any natural representation. You can assume a higher score
indicates higher reliability in the cluster's content, but this is
only partially accurate (for example, the score incorporates the
cluster's size in STC).

Cluster scores need not be normalized, although for STC and Lingo I
think they are (would have to check the code). You can always do
post-processing normalization (min-max).

> Can I get what is called the label score? if yes, how? what does it represents?

Cluster score and label score are equivalent. I don't know what you mean.

> how can I set the Label_Similarity_Threshold LINGO parameter?

For setting attributes see UsingAttributes.java example:

https://fisheye3.atlassian.com/browse/carrot2/trunk/applications/carrot2-examples/src/org/carrot2/examples/clustering/UsingAttributes.java#r4561

But I don't know which attribute you have in mind -- can you point at
the source code or documentation where you encountered this attribute?

Dawid

------------------------------------------------------------------------------
Beautiful is writing same markup. Internet Explorer 9 supports
standards for HTML5, CSS3, SVG 1.1,  ECMAScript5, and DOM L2 & L3.
Spend less time writing and  rewriting code and more time creating great
experiences on the web. Be a part of the beta today
http://p.sf.net/sfu/msIE9-sfdev2dev
_______________________________________________
Carrot2-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/carrot2-developers
Reply | Threaded
Open this post in threaded view
|

Re: Cluster Scores & similarity

Stanislaw Osinski
Administrator
> how can I set the Label_Similarity_Threshold LINGO parameter?

For setting attributes see UsingAttributes.java example:

https://fisheye3.atlassian.com/browse/carrot2/trunk/applications/carrot2-examples/src/org/carrot2/examples/clustering/UsingAttributes.java#r4561

But I don't know which attribute you have in mind -- can you point at
the source code or documentation where you encountered this attribute?

I guess you meant an attribute that was there in the original algorithm (described in the paper and available in Carrot2 2.x). Lingo has been updated since that time, and that attribute got replaced with a number of others. Take a look at this thread for more information:


S.


------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Carrot2-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/carrot2-developers
Reply | Threaded
Open this post in threaded view
|

Re: Cluster Scores & similarity

mounaim.latif
how the score of  cluster is calculated ?
Reply | Threaded
Open this post in threaded view
|

Re: Cluster Scores & similarity

Dawid Weiss-2
The source code is in the repository, Mounaim. These metrics are
fairly standard in IR so any book (or web search) should bring you a
ton of references.

Dawid

On Thu, Jun 14, 2012 at 2:40 AM, mounaim.latif <[hidden email]> wrote:

> how the score of  cluster is calculated ?
>
> --
> View this message in context: http://carrot2-users-and-developers-forum.607571.n2.nabble.com/Cluster-Scores-similarity-tp5763175p7577544.html
> Sent from the Carrot2 Users and Developers Forum mailing list archive at Nabble.com.
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Carrot2-developers mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/carrot2-developers
>

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Carrot2-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/carrot2-developers