Published August 25, 2021
| Version v1
Dataset
Open
TREC
Authors/Creators
Description
TREC with 5,952 documents (i.e. questions), is a question classification dataset in which the task is classify a question into 6 main subject categories: such as human, location, entity, abbreviation, description and numeric value.
The files:
texts.txt: Document set (text). One per line.
score.txt: Document class whose index is associated with texts.txt
split_<k>.pkl: pandas DataFrame with k-cross validation partition