Re: Cluster analysis

hi, this is Nitish. I am working as a tec consultant for Statistica
software in a company called Statsoft India. I think, your purpose will
be solved in statistica. You can make use of cluster anaysis option of
multivariate analysis module of statistica. In which, you will be
having two methods of cluster analysis: tree joining and k-means
clustering. Tree joining method gives you an idea that how many
possiable clusters can be created for your data. Once you get an idea
about the number of clusters, you can go for k-means clustering method
to cluster your data. This method allows you to create user-defined no
of clusters for your data and view the properties of different clusters
in terms of cluster means and means of different variables of your
dataset in different clusters.
Apart of this, you have a data mining tool called Statistica Dataminer,
which can again be used for cluster analysis.In Statistica Data
Miner,you will be having neural networks and Generalized EM and k-Means
Cluster Analysis.Both of these tools can be used for cluster analysis.
The best part about these tools is that you dont have to specify the no
of clusters you will be interested in creating. The best cluster
solution for your data will be given by Statistica Dataminer on its
own.the Generalized EM and k-Means Cluster Analysis module uses a
modified v-fold cross-validation scheme to determine the best number
of clusters from the data.Statistica can take care of both the
categorial and continous types of variable for cluster analysis.
I am just suggesting you one of the tools which can be used for cluster
analysis.
I hope this will help.

Regards,
Nitish
Eric wrote:
I would like to perform a cluster analysis on initially three variables
but possible extended to more if the initial analysis produces "good
results". Of the 3 variables, 1 is a continuous variable and the other
two are discrete variables and I'm using S-Plus and R to perform the
cluster analysis. I would like to construct either 2 or 3 clusters from
the analysis, but I'm not sure which method I should be using and the
metric. Can anyone one suggest the most appropriate method and metric
given the variables?

I would be very grateful for any asistence on this topic.


Best regards,

Eric

.

Relevant Pages

  • Re: Finding useful functions- part 1
    ... Bear in mind that I actualy write cluster ... > The one thing one soon learns after using cluster analysis practically ... approach to the functional classification of neurons, ... cells can be clustered so that most of the diversity is captured by the ...
    (sci.cognitive)
  • Re: clustering
    ... > going from a five cluster step to a 4 cluster step, ... into the subject of "cluster analysis" about a DECADE after I ... ALL existing clustering algorithms! ... In Art' second sentence, it is imperative to drop the word "easily". ...
    (sci.stat.edu)
  • Re: Clustering categorical data
    ... cluster analysis, ... your "scales" are internally symmetrical in wording, ... objects/cases here are "items", which are measured on ...
    (sci.stat.math)
  • Re: Cluster analysis on dataset with ordinal and nominal data
    ... The TWOSTEP cluster procedure in SPSS handles variables at different ... If you are using Likert scales, ... > 2) Is there any alternative method of cluster analysis of this dataset ...
    (sci.stat.edu)