Datasets:
Commit
·
bc4eb06
1
Parent(s):
5b73d27
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,55 +1,17 @@
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
- expert-generated
|
| 6 |
language:
|
| 7 |
- en
|
| 8 |
-
|
| 9 |
-
-
|
| 10 |
-
|
| 11 |
-
-
|
|
|
|
| 12 |
pretty_name: PET
|
| 13 |
size_categories:
|
| 14 |
-
-
|
| 15 |
-
source_datasets:
|
| 16 |
-
- Friedrich et al. original dataset
|
| 17 |
-
task_categories:
|
| 18 |
-
- token-classification
|
| 19 |
-
task_ids:
|
| 20 |
-
- token classification
|
| 21 |
-
- named entity recognition
|
| 22 |
-
- relation extraction
|
| 23 |
-
dataset_info:
|
| 24 |
-
features:
|
| 25 |
-
- name: document name
|
| 26 |
-
dtype: string
|
| 27 |
-
- name: tokens
|
| 28 |
-
sequence: string
|
| 29 |
-
- name: tokens-IDs
|
| 30 |
-
sequence: int8
|
| 31 |
-
- name: ner_tags
|
| 32 |
-
sequence: string
|
| 33 |
-
- name: sentence-IDs
|
| 34 |
-
sequence: int8
|
| 35 |
-
- name: relations
|
| 36 |
-
sequence:
|
| 37 |
-
- name: source-head-sentence-ID
|
| 38 |
-
dtype: int8
|
| 39 |
-
- name: source-head-word-ID
|
| 40 |
-
dtype: int8
|
| 41 |
-
- name: relation-type
|
| 42 |
-
dtype: string
|
| 43 |
-
- name: target-head-sentence-ID
|
| 44 |
-
dtype: int8
|
| 45 |
-
- name: target-head-word-ID
|
| 46 |
-
dtype: int8
|
| 47 |
-
splits:
|
| 48 |
-
- name: test
|
| 49 |
-
num_bytes: 203379
|
| 50 |
-
num_examples: 45
|
| 51 |
-
download_size: 38326
|
| 52 |
-
dataset_size: 203379
|
| 53 |
---
|
| 54 |
|
| 55 |
# PET: A NEW DATASET FOR PROCESS EXTRACTION FROM TEXT
|
|
@@ -288,24 +250,56 @@ The *Further Specification* process element was the most difficult element to an
|
|
| 288 |
### Citation Information
|
| 289 |
|
| 290 |
```
|
| 291 |
-
@
|
| 292 |
-
author
|
| 293 |
-
|
| 294 |
-
|
| 295 |
-
|
| 296 |
-
|
| 297 |
-
|
| 298 |
-
|
| 299 |
-
|
| 300 |
-
|
| 301 |
-
|
| 302 |
-
|
| 303 |
-
|
| 304 |
-
|
| 305 |
-
|
| 306 |
-
|
| 307 |
-
|
| 308 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 309 |
}
|
| 310 |
```
|
| 311 |
|
|
@@ -318,7 +312,7 @@ Thanks to [Patrizio Bellan](https://pdi.fbk.eu/bellan/) for adding this dataset.
|
|
| 318 |
#### <a name="updates"></a>Update
|
| 319 |
- v1.0.0: Added token classification task
|
| 320 |
- v1.0.1: Added extraction relation task
|
| 321 |
-
|
| 322 |
|
| 323 |
|
| 324 |
## <a name="annotationguidelines"></a>Annotation Guidelines
|
|
@@ -341,7 +335,7 @@ A python interface (beta version) to interact with the dataset can be found [her
|
|
| 341 |
|
| 342 |
### Benchmarks
|
| 343 |
|
| 344 |
-
A python benchmarking procedure to test approaches on the PET dataset
|
| 345 |
|
| 346 |
|
| 347 |
## <a name="loadingdata"></a>Loading data
|
|
|
|
| 1 |
---
|
| 2 |
+
license: mit
|
| 3 |
+
task_categories:
|
| 4 |
+
- token-classification
|
|
|
|
| 5 |
language:
|
| 6 |
- en
|
| 7 |
+
tags:
|
| 8 |
+
- Business Process Management
|
| 9 |
+
- NLP
|
| 10 |
+
- ML
|
| 11 |
+
- DL
|
| 12 |
pretty_name: PET
|
| 13 |
size_categories:
|
| 14 |
+
- n<1K
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
---
|
| 16 |
|
| 17 |
# PET: A NEW DATASET FOR PROCESS EXTRACTION FROM TEXT
|
|
|
|
| 250 |
### Citation Information
|
| 251 |
|
| 252 |
```
|
| 253 |
+
@inproceedings{DBLP:conf/aiia/BellanGDPA22,
|
| 254 |
+
author = {Patrizio Bellan and
|
| 255 |
+
Chiara Ghidini and
|
| 256 |
+
Mauro Dragoni and
|
| 257 |
+
Simone Paolo Ponzetto and
|
| 258 |
+
Han van der Aa},
|
| 259 |
+
editor = {Debora Nozza and
|
| 260 |
+
Lucia C. Passaro and
|
| 261 |
+
Marco Polignano},
|
| 262 |
+
title = {Process Extraction from Natural Language Text: the {PET} Dataset and
|
| 263 |
+
Annotation Guidelines},
|
| 264 |
+
booktitle = {Proceedings of the Sixth Workshop on Natural Language for Artificial
|
| 265 |
+
Intelligence {(NL4AI} 2022) co-located with 21th International Conference
|
| 266 |
+
of the Italian Association for Artificial Intelligence (AI*IA 2022),
|
| 267 |
+
Udine, November 30th, 2022},
|
| 268 |
+
series = {{CEUR} Workshop Proceedings},
|
| 269 |
+
volume = {3287},
|
| 270 |
+
pages = {177--191},
|
| 271 |
+
publisher = {CEUR-WS.org},
|
| 272 |
+
year = {2022},
|
| 273 |
+
url = {https://ceur-ws.org/Vol-3287/paper18.pdf},
|
| 274 |
+
timestamp = {Fri, 10 Mar 2023 16:23:01 +0100},
|
| 275 |
+
biburl = {https://dblp.org/rec/conf/aiia/BellanGDPA22.bib},
|
| 276 |
+
bibsource = {dblp computer science bibliography, https://dblp.org}
|
| 277 |
+
}
|
| 278 |
+
|
| 279 |
+
@inproceedings{DBLP:conf/bpm/BellanADGP22,
|
| 280 |
+
author = {Patrizio Bellan and
|
| 281 |
+
Han van der Aa and
|
| 282 |
+
Mauro Dragoni and
|
| 283 |
+
Chiara Ghidini and
|
| 284 |
+
Simone Paolo Ponzetto},
|
| 285 |
+
editor = {Cristina Cabanillas and
|
| 286 |
+
Niels Frederik Garmann{-}Johnsen and
|
| 287 |
+
Agnes Koschmider},
|
| 288 |
+
title = {{PET:} An Annotated Dataset for Process Extraction from Natural Language
|
| 289 |
+
Text Tasks},
|
| 290 |
+
booktitle = {Business Process Management Workshops - {BPM} 2022 International Workshops,
|
| 291 |
+
M{\"{u}}nster, Germany, September 11-16, 2022, Revised Selected
|
| 292 |
+
Papers},
|
| 293 |
+
series = {Lecture Notes in Business Information Processing},
|
| 294 |
+
volume = {460},
|
| 295 |
+
pages = {315--321},
|
| 296 |
+
publisher = {Springer},
|
| 297 |
+
year = {2022},
|
| 298 |
+
url = {https://doi.org/10.1007/978-3-031-25383-6\_23},
|
| 299 |
+
doi = {10.1007/978-3-031-25383-6\_23},
|
| 300 |
+
timestamp = {Tue, 14 Feb 2023 09:47:10 +0100},
|
| 301 |
+
biburl = {https://dblp.org/rec/conf/bpm/BellanADGP22.bib},
|
| 302 |
+
bibsource = {dblp computer science bibliography, https://dblp.org}
|
| 303 |
}
|
| 304 |
```
|
| 305 |
|
|
|
|
| 312 |
#### <a name="updates"></a>Update
|
| 313 |
- v1.0.0: Added token classification task
|
| 314 |
- v1.0.1: Added extraction relation task
|
| 315 |
+
- v1.1.0: Fixed minor errors, fixed performs relations
|
| 316 |
|
| 317 |
|
| 318 |
## <a name="annotationguidelines"></a>Annotation Guidelines
|
|
|
|
| 335 |
|
| 336 |
### Benchmarks
|
| 337 |
|
| 338 |
+
A python benchmarking procedure package to test approaches on the PET dataset ca be found [here](https://pypi.org/project/petbenchmarks/)
|
| 339 |
|
| 340 |
|
| 341 |
## <a name="loadingdata"></a>Loading data
|