pg.ddx.io  pgsql-students@postgresql.org mailing list archive  
help / color / mirror / Atom feed
: Fuzzy C means algorithms in MAdlib
2+ messages / 2 participants
[nested] [flat]

* : Fuzzy C means algorithms in MAdlib
@ 2013-04-27 10:36  Akansha Singh <akansha.singh@oracle.com>
  0 siblings, 1 reply; 2+ messages in thread

From: Akansha Singh @ 2013-04-27 10:36 UTC (permalink / raw)
  To: pgsql-students

HI
 I would like to know can this algorithm be implemented on MAdlIB
In the K-means algorithm, each vector is classified as belonging to a single cluster (hard clustering), and the centroids are updated based on the classified samples. In a variation of this approach known as fuzzy c-means, all vectors have a degree of membership for each cluster, and the respective centroids are calculated based on these membership degrees.

Whereas the K-means algorithm computes the average of the vectors in a cluster as the center, fuzzy c-means finds the center as a weighted average of all points, using the membership probabilities for each point as weights. Vectors with a high probability of belonging to the class have larger weights, and more influence on the centroid.

Regards
Akansha Singh


-- 
Sent via pgsql-students mailing list (pgsql-students@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-students


-- 
Sent via pgsql-students mailing list (pgsql-students@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-students



^ permalink  raw  reply  [nested|flat] 2+ messages in thread

* Re: : Fuzzy C means algorithms in MAdlib
@ 2013-05-01 22:04  Tomas Vondra <tv@fuzzy.cz>
  parent: Akansha Singh <akansha.singh@oracle.com>
  0 siblings, 0 replies; 2+ messages in thread

From: Tomas Vondra @ 2013-05-01 22:04 UTC (permalink / raw)
  To: pgsql-students

On 27.4.2013 12:36, Akansha Singh wrote:
> HI I would like to know can this algorithm be implemented on MAdlIB 
> In the K-means algorithm, each vector is classified as belonging to a
> single cluster (hard clustering), and the centroids are updated based
> on the classified samples. In a variation of this approach known as
> fuzzy c-means, all vectors have a degree of membership for each
> cluster, and the respective centroids are calculated based on these
> membership degrees.
> 
> Whereas the K-means algorithm computes the average of the vectors in
> a cluster as the center, fuzzy c-means finds the center as a weighted
> average of all points, using the membership probabilities for each
> point as weights. Vectors with a high probability of belonging to the
> class have larger weights, and more influence on the centroid.

While I'm a fan of data analysis, I'm still struggling with a question
why this should be implemented as a PostgreSQL GSoC project. What would
be the result? An update to MADlib, implementing k-means, or somethink
like a PostgreSQL extension?

regards
Tomas


-- 
Sent via pgsql-students mailing list (pgsql-students@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-students



^ permalink  raw  reply  [nested|flat] 2+ messages in thread


end of thread, other threads:[~2013-05-01 22:04 UTC | newest]

Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2013-04-27 10:36 : Fuzzy C means algorithms in MAdlib Akansha Singh <akansha.singh@oracle.com>
2013-05-01 22:04 ` Tomas Vondra <tv@fuzzy.cz>

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox