hsgsom

An implementation of the GSOM clustering algorithm.

Latest on Hackage:0.2.0

This package is not currently in any snapshots. If you're interested in using it, we recommend adding it to Stackage Nightly. Doing so will make builds more reliable, and allow stackage.org to host generated Haddocks.

BSD3 licensed by Stephan Günther
Maintained by Stephan Günther <gnn dot github at gmail dot com>
= README

This is the README file of hsgsom, a haskell library implementing the
growing self organising map clustering algorithm.

== The Algorithm

The growing self organising map (GSOM) algorithm is a clustering algorithm
working on a set of n-dimensional numeric input vectors. It's output is a
network of nodes laid out in two dimensions where each node has a weight
vector associated with it. This weight vector has the same dimension as the
input vectors and is meant to be intepreted as a cluster center, i.e. it
represents those input vectors whose distance to the node's weight vector
is minimal when compared to the distance to the other nodes weight vectors.
See <http://en.wikipedia.org/wiki/GSOM> for an explanation of the algorithm.
The algorithm was introduced in:

Alahakoon, D., Halgamuge, S. K. and Sirinivasan, B. (2000)
Dynamic Self Organizing Maps With Controlled Growth
for Knowledge Discovery,
IEEE Transactions on Neural Networks,
Special Issue on Knowledge Discovery and Data Mining, 11, pp 601-614.

== License

This package and its contents are licensed under the BSD 3 clause license.
You should have received a file called LICENSE containing said license
along with this package.

== Versioning

This README correponds to version 0.1.0 of hsgsom, so as you can see it is
a very early version.
Version numbers follow the pattern X.Y.Z and have the following meaning:

- a change in Z corresponds to minor changes as in documentation changes
or changes to the underlying implementation

- a change in Y correponds to added functionality and/or backwards
compatible interface changes/additions.

- a change in X correpsonds to a major implementation change either
drastically changing the algorithm behaviour or performance or
changing the interface in a possibly not backwards compatible way.

== Questions, Bugs, etc...

If you think you have found a bug, or you have questions or suggestions
or really anything to say about the package it would be greatly appreciated
if you would drop me a note or an email.
This is my very firt attempt at packaging and releasing a substantial amount
my own code to the public and I'm eager to learn how to do thinks better.

Thanks for using/looking at this package and have a nice day.
Depends on:
Used by 1 package:
comments powered byDisqus