Cluster Analysis:build a classifier?

classic Classic list List threaded Threaded
3 messages Options
x_x
Reply | Threaded
Open this post in threaded view
|

Cluster Analysis:build a classifier?

x_x
Dear All,
I'm currently doing a project about unsupervised learning, and I'll be using R to analyse a few network traffic datasets downloaded off Andrew Moore's website (http://www.cl.cam.ac.uk/Research/SRG/netos/nprobe/data/papers/sigmetrics/index.html).

Could anyone shed some light on how to build a classifier from the training set A, and then use a validation set B to assess the performance of the classifier? I'm able to group these datasets into groups using hclust() and kmeans() etc., but I don't see the connection between the clustered dataset A and the remaining sets.

Any thoughts would be much appreciated!

Yours sincerely,
x_x
Reply | Threaded
Open this post in threaded view
|

Re: Cluster Analysis:build a classifier?

Uwe Ligges
x_x wrote:

> Dear All,
> I'm currently doing a project about unsupervised learning, and I'll be using
> R to analyse a few network traffic datasets downloaded off Andrew Moore's
> website
> (http://www.cl.cam.ac.uk/Research/SRG/netos/nprobe/data/papers/sigmetrics/index.html).
>
> Could anyone shed some light on how to build a classifier from the training
> set A, and then use a validation set B to assess the performance of the
> classifier? I'm able to group these datasets into groups using hclust() and
> kmeans() etc., but I don't see the connection between the clustered dataset
> A and the remaining sets.
>
> Any thoughts would be much appreciated!
>
> Yours sincerely,
> x_x

Dear x_x,

you are confusing unsupervised learning/clustering with supervised
learning/classification, I fear. Please ask your teacher/supervisor for
getting the terminology right. Please note that it is good practice to
sign e-mail messages with your real name.

Best,
Uwe Ligges

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
x_x
Reply | Threaded
Open this post in threaded view
|

Re: Cluster Analysis:build a classifier?

x_x
Dear Uwe,
Thank you very much for your reply, I think that's why I've done lots of research on this to no avail. And also, thanks for the suggestion on the signature.

Best wishes,
Yang
Uwe Ligges wrote
x_x wrote:
> Dear All,
> I'm currently doing a project about unsupervised learning, and I'll be using
> R to analyse a few network traffic datasets downloaded off Andrew Moore's
> website
> (http://www.cl.cam.ac.uk/Research/SRG/netos/nprobe/data/papers/sigmetrics/index.html).
>
> Could anyone shed some light on how to build a classifier from the training
> set A, and then use a validation set B to assess the performance of the
> classifier? I'm able to group these datasets into groups using hclust() and
> kmeans() etc., but I don't see the connection between the clustered dataset
> A and the remaining sets.
>
> Any thoughts would be much appreciated!
>
> Yours sincerely,
> x_x

Dear x_x,

you are confusing unsupervised learning/clustering with supervised
learning/classification, I fear. Please ask your teacher/supervisor for
getting the terminology right. Please note that it is good practice to
sign e-mail messages with your real name.

Best,
Uwe Ligges

______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.