multi-labeled samples

[Ferhat KURTULMUŞ]

I don't know if it is a proper place to ask my question. I have data with multi-labeled samples. Namely, my data has instances which may belong to two independent classes in the same time (not multi-classes). My question raises the point that I want to perform one or more methods (eg., stepwise discriminant analysis) to eliminate ineffective features. And I wonder if the milk library supports multi-labeling for both stepdisc and training-testing with any classifier. One think that I can consider is performing stepwise discriminant analysis twice using different label vectors corresponding to each class groups, and creating a filtered data based on common responses of two stepdisc analysis (maybe concurring features).

Thanks in advance,

Ferhat

[Luis Pedro Coelho]

On 11/26/2013 10:58 AM, Ferhat KURTULMUŞ wrote:
> I don't know if it is a proper place to ask my question.

Yes, it is.

> I have data
> with multi-labeled samples

> <http://en.wikipedia.org/wiki/Multi-label_classification>. Namely, my

> data has instances which may belong to two independent classes in the
> same time (not multi-classes). My question raises the point that I want
> to perform one or more methods (eg., stepwise discriminant analysis) to
> eliminate ineffective features. And I wonder if the milk library
> supports multi-labeling for both stepdisc and training-testing with any
> classifier. One think that I can consider is performing stepwise
> discriminant analysis twice using different label vectors corresponding
> to each class groups, and creating a filtered data based on common
> responses of two stepdisc analysis (maybe concurring features).

Milk only supports a very simplistic model where it learns separate
binary models for each possible labels and, when classifying, outputs
all those that match.

here is the code to convert a learner (binary or multi-class learner)
into a multi_label learner:

import milk
from milk.supervised.multi_label import one_by_one
learner = milk.defaultlearner()
learner = one_by_one(learner)


HTH,
--
Luis Pedro Coelho | EMBL | http://luispedro.org

Recent stuff:
http://dx.doi.org/10.5334/jors.ac
http://bit.ly/coelho2013-video
http://bit.ly/general-subcellular-determination

[Ferhat KURTULMUŞ]

I appreciate it for your valuable answer. I will give it a try.

27 Kasım 2013 Çarşamba 09:59:13 UTC+2 tarihinde Luis Pedro Coelho yazdı: