Datasets:

josephimperial
/

CebuaNER

Modalities:

Text

Formats:

text

Size:

100K - 1M

ArXiv:

Libraries:

Datasets

License:

Dataset card Viewer Files Files and versions Community

Dataset Viewer

Auto-converted to Parquet

Full Screen Viewer

Full Screen

Split (1)

train · 188k rows

text stringlengths 0 38
-DOCSTART- -X- O
Aduna'y -X- _ O
pagsaka -X- _ O
sa -X- _ O
presyo -X- _ O
sa -X- _ O
lana -X- _ O
, -X- _ O
epektibo -X- _ O
sa -X- _ O
Martes -X- _ O
, -X- _ O
Hunyo -X- _ O
27 -X- _ O
, -X- _ O
2023 -X- _ O
. -X- _ O

MOTORISTA -X- _ O
, -X- _ O
PATAY -X- _ O
HUMAN -X- _ O
MALIGSAN -X- _ O
ANG -X- _ O
ULO -X- _ O
SA -X- _ O
10-WHEELER -X- _ O
DUMP -X- _ O
TRUCK -X- _ O
Patay -X- _ O
ang -X- _ O
usa -X- _ O
ka -X- _ O
motorista -X- _ O
sa -X- _ O
usa -X- _ O
ka -X- _ O
vehicular -X- _ O
accident -X- _ O
ang -X- _ O
nahitabo -X- _ O
sa -X- _ O
Barangay -X- _ B-LOC
Caidiocan -X- _ I-LOC
sa -X- _ O
Valencia -X- _ B-LOC
mga -X- _ O
alas-6:20 -X- _ O
sa -X- _ O
gabii -X- _ O
niadtong -X- _ O
Sabado -X- _ O
, -X- _ O
Hunyo -X- _ O
24,2023. -X- _ O
Giila -X- _ O
ang -X- _ O
biktima -X- _ O
nga -X- _ O
si -X- _ O
Ruperto -X- _ B-PER
Solamillo -X- _ I-PER
Ausejo -X- _ I-PER
Jr. -X- _ I-PER
, -X- _ O
46 -X- _ O
anyos -X- _ O
, -X- _ O
ulitawo -X- _ O
, -X- _ O
ug -X- _ O
lumolupyo -X- _ O
sa -X- _ O
Barangay -X- _ B-LOC
Pulang -X- _ I-LOC
Bato -X- _ I-LOC
sa -X- _ O
naasoy -X- _ O
nga -X- _ O
lungsod. -X- _ O
Sumala -X- _ O
pa -X- _ O
sa -X- _ O
imbestigasyon -X- _ O
sa -X- _ O
kapulisan -X- _ O
sa -X- _ O
Valencia -X- _ B-LOC
, -X- _ O
nagbiyahe -X- _ O
ang -X- _ O
duha -X- _ O
sa -X- _ O
managlahing -X- _ O
direksyon. -X- _ O
Pag-abot -X- _ O
sa -X- _ O
nahisgutang -X- _ O
lugar -X- _ O
, -X- _ O

End of preview. Expand in Dataset Viewer.

This repository contains CebuaNER, the largest gold-standard datasets for named entities in Cebuano. This dataset is used for the paper CebuaNER: A New Baseline Cebuano Named Entity Recognition Model to be presented at PACLIC 2023, authored by Ma. Beatrice Emanuela N. Pilar, Ellyza Mari J. Papas, Mary Loise Buenaventura, Dane C. Dedoroy, Myron Montefalcon, Jay Rhald Padilla, Lany Maceda, Mideth Abisado, and Joseph Imperial.

Data

The dataset contribution of this study is a compilation of Cebuano news articles from two local news stations, Yes the Best Dumaguete and the Filipinas Bisaya. To further increase the data count, we also incorporated another publicly available dataset from Sunstar Cebu pre-collected by independent researcher Arjemariel Requina. The total accumulated and filtered size of the Cebuano dataset is 4258 articles. All the dataset used for this study is purely for non-commercial research purposes and under Fair Use.

Source	Original	Cleaned
Yes the Best Dumaguete	1484	781
Filipinas Bisaya	769	377
Sunstar	3100	3100

Note on Data Cataloging

Please send an email before submitting this repository to any data cataloging, data aggregation, and benchmarking projects/initiatives. The proponents of the paper of this dataset would like to be acknowledged appropriately or involved in co-authorship.

Contact

For questions, you may reach the corresponding author of the study at:

Joseph Marvin Imperial
Faculty Member, Department of Computer Science
Lab Head, NU Human Language Technology Lab
jrimperial@national-u.edu.ph

Downloads last month: 36

Size of downloaded dataset files:

2.63 MB

Size of the auto-converted Parquet files:

1.13 MB

Number of rows:

188,291