Datasets:
Tasks:
Text Classification
Formats:
parquet
Sub-tasks:
multi-label-classification
Languages:
English
Size:
100K - 1M
License:
Question
#2
by
akkasi
- opened
Hi,
What are the labels to be predicted in this dataset if I want to do multi-label classification?
tnx
I had the same question. Indeed, it seems there are no obvious labels for prediction. So, I looked up the original OHSUMED dataset, which categorizes all abstracts into 23 cardiovascular diseases, a detail I couldn't find in this dataset card. I'm confused too.
And for ur information, here's the website of the original dataset: https://disi.unitn.it/moschitti/corpora.htm