Published June 28, 2020
| Version v2
Dataset
Open
Weight 20Newsgroup words.
Authors/Creators
Description
Subset of the data used in "Quantitative Evaluation of Machine Learning Explanations: A Human-Grounded Benchmark". Contains words from documents in the 20Newsgroup dataset, which were weighted based on annotators' selection for each text article. Human-attention masks for text samples are in JSON format files.
Files
20newsgroup.zip
Files
(290.5 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:3362500ad91bfbd2626f70ac68c63e93
|
290.5 kB | Preview Download |