Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
Datasets:
redasers
/
difraud
like
2
Follow
ReDAS Lab at the University of Houston
5
Tasks:
Text Classification
Zero-Shot Classification
Languages:
English
Size:
10K<n<100K
Tags:
fraud-detection
deception-detection
phishing
fake-news
benchmark
opinion-spam
+ 1
License:
mit
Dataset card
Files
Files and versions
Community
d7450fa
difraud
/
README.md
Commit History
redone yaml following GLUE
d7450fa
dainis-boumber
commited on
Sep 29, 2023
yaml fix
3b77f86
dainis-boumber
commited on
Sep 28, 2023
cleanup
45bc599
dainis-boumber
commited on
Sep 28, 2023
Fixed appendix
97949b0
dainis-boumber
commited on
Sep 28, 2023
updated dataset descriptions
a7a8c3b
dainis-boumber
commited on
Sep 28, 2023
fixed numbers
64b9954
dainis-boumber
commited on
Sep 28, 2023
Updated readme
f95872a
dainis-boumber
commited on
Sep 28, 2023
added authors
04fbdc7
dainis-boumber
commited on
Sep 28, 2023
Update README.md
4db0676
dainis-boumber
commited on
Sep 28, 2023
create dataset
cd17ca5
dainis-boumber
commited on
Sep 28, 2023
initial commit
141fdd2
dainis-boumber
commited on
Sep 28, 2023