Blobs in ARFF files

[Marcin Wojnarski]

Hi,

Question to Peter. Is it possible to keep binary objects (BLOBs), like
images, in ARFF files? Or do you plan to add such a feature, for
instance as a new type of attribute?

Thanks,
Marcin

--
Marcin Wojnarski, Project Lead, TunedIT, http://tunedit.org
Machine Learning & Data Mining Research -
Automated Tests, Repeatable Experiments, Meaningful Results


_______________________________________________
Wekalist mailing list
Send posts to: Wekalist@...
List info and subscription status: https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

[Peter Reutemann-3]

> Question to Peter. Is it possible to keep binary objects (BLOBs), like
> images, in ARFF files? Or do you plan to add such a feature, for instance as
> a new type of attribute?

I'm not the Weka maintainer (this is Mark Hall at the moment), I'm
just volunteering on the mailing list. So far, I haven't heard of any
such plans, but I'll pass this question on to him. ;-)

Cheers, Peter
--
Peter Reutemann, Dept. of Computer Science, University of Waikato, NZ
http://www.cs.waikato.ac.nz/~fracpete/           Ph. [phone]

_______________________________________________
Wekalist mailing list
Send posts to: Wekalist@...
List info and subscription status: https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

[Mark Hall-9]

On 27/10/09 11:42 PM, Peter Reutemann wrote:
>> Question to Peter. Is it possible to keep binary objects (BLOBs), like
>> images, in ARFF files? Or do you plan to add such a feature, for instance as
>> a new type of attribute?
>
> I'm not the Weka maintainer (this is Mark Hall at the moment), I'm
> just volunteering on the mailing list. So far, I haven't heard of any
> such plans, but I'll pass this question on to him. ;-)

There are no such plans at present. Derived attributes would have to be created
from BLOBs before any standard learners could be applied to such data anyway.

Cheers,
Mark.


_______________________________________________
Wekalist mailing list
Send posts to: Wekalist@...
List info and subscription status: https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html