Re: Data Metrics

Scott Dick (dick@morden.csee.usf.edu)
Sun, 26 Jul 1998 16:09:04 +0200 (MET DST)

--------------AE5470448B1D60BFEFB677A9
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hello,

I did something like this in an upcoming paper. Think of each record as a
multidimensional vector, where each field of the record is a dimension. If
the fields are numeric, you're basically done. If the fields are not numeric,
then you need to do some more work. Can you impose an ordering on the
possible values of the fields? If so, you can use the number of values
appearing between a field value and a reference value (a "zero" value) as a
distance measure. The idea appears in a different context in

Stilman, B. "Linguistic Geometry: Methodology and Techniques". In Cybernetics
and Systems, Vol. 26, 1995, pp 535-597.

Hope this helps!

Camilo wrote:

> Hi,
> does anyone know about approaches to theories on the definition of
> a metric for data?
>
> Something like measuring the "distance" between two records, based
> on the data they contain, NOT on the repository (their "physical" distance
> on a hard disk plate, which wouldn't be that sensical anyway).
>
> Please, I know this is an uncharted terrain, so communication is not so
> straight forward.
> Telling me ... "use fuzzy logic".. when a metric theory approach hasn't
> been defined for a fuzzy set is not helpful.
>
> If you have thought about it, what could mean it?
>
> Thanks
>
> Camilo

--
***********************************************************************
*                               *                                     *
*  Scott Dick                   *                                     *
*  Research Assistant           *       Cool & brilliant thought      *
*  USF Computer Science         *       still under contruction       *
*  dick@morden.csee.usf.edu     *                                     *
*                               *                                     *
***********************************************************************

--------------AE5470448B1D60BFEFB677A9 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit

Hello,

    I did something like this in an upcoming paper. Think of each record as a multidimensional vector, where each field of the record is a dimension. If the fields are numeric, you're basically done. If the fields are not numeric, then you need to do some more work. Can you impose an ordering on the possible values of the fields? If so, you can use the number of values appearing between a field value and a reference value (a "zero" value) as a distance measure. The idea appears in a different context in

Stilman, B. "Linguistic Geometry: Methodology and Techniques". In Cybernetics and Systems, Vol. 26, 1995, pp 535-597.

Hope this helps!
 

Camilo wrote:

 Hi,
    does anyone know about approaches to theories on the definition of
a metric for data?

 Something like measuring the "distance" between two records, based
on the data they contain, NOT on the repository (their "physical" distance
on a hard disk plate, which wouldn't be that sensical anyway).

Please, I know this is an uncharted terrain, so communication is not so
straight forward.
 Telling me ...  "use fuzzy logic".. when a metric theory approach hasn't
been defined for a fuzzy set is not helpful.

 If you have thought about it, what could mean it?

 Thanks

 Camilo

 
-- 
***********************************************************************
*                               *                                     *
*  Scott Dick                   *                                     *
*  Research Assistant           *       Cool & brilliant thought      *         
*  USF Computer Science         *       still under contruction       *
*  dick@morden.csee.usf.edu     *                                     *
*                               *                                     *
***********************************************************************
  --------------AE5470448B1D60BFEFB677A9-- ############################################################################ This message was posted through the fuzzy mailing list. (1) To subscribe to this mailing list, send a message body of "SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at (2) To unsubscribe from this mailing list, send a message body of "UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com" to listproc@dbai.tuwien.ac.at (3) To reach the human who maintains the list, send mail to fuzzy-owner@dbai.tuwien.ac.at (4) WWW access and other information on Fuzzy Sets and Logic see http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info (5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html