Re: PGSQL: Fuzzy C means algorithms in MAdlib

Sent from my iPad

On 27-Apr-2013, at 16:04, Akansha Singh <akansha(dot)singh(at)oracle(dot)com> 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 [2, 29], 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.

I would suggest marking the MADlib devel list as well.

This seems like an easy to do variation of k means, and should be doable and built over the existing k means resources in MADLib.

Regards,

Atri