Train data

42 views

Skip to first unread message

yhh...@nlg.csie.ntu.edu.tw

unread,

Sep 8, 2016, 8:20:31 AM9/8/16

to SemEval-2017 Task 7: Detection and Interpretation of English Puns

Dear organizer,

Thanks for this interesting task!

I wonder that when will the train data ready?

I can only find trial data now.

Thanks,

Victor

Tristan Miller

unread,

Sep 8, 2016, 9:01:47 AM9/8/16

to semeva...@googlegroups.com

Dear Victor,

On 08/09/16 02:20 PM, yhh...@nlg.csie.ntu.edu.tw wrote:
> Thanks for this interesting task!
> I wonder that when will the train data ready?
> I can only find trial data now.

Thanks for your interest in the task. Unlike in traditional lexical
sample WSD, it is extremely difficult to amass a large number of pun
examples per word or per sense. For this reason, we will be providing
only trial and test data, but no training data. It is expected that
this shared task is most amenable to approaches that are unsupervised or
knowledge-based rather than supervised.

I thought we had mentioned this on the task website but I see now that
I'd overlooked that. I'll update it soon with an explanation.

Regards,
Tristan

--
Tristan Miller, Research Scientist
Ubiquitous Knowledge Processing Lab (UKP-TUDA)
Department of Computer Science, Technische Universität Darmstadt
Tel: +49 6151 162 5296 | Web: https://www.ukp.tu-darmstadt.de/