Datasets:
Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
named-entity-recognition
Size:
< 1K
License:
Update README.md
Browse files
README.md
CHANGED
@@ -38,6 +38,35 @@ pretty_name: SwissNER
|
|
38 |
size_categories:
|
39 |
- n<1K
|
40 |
---
|
41 |
-
# Dataset Card for "swissner"
|
42 |
|
43 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
size_categories:
|
39 |
- n<1K
|
40 |
---
|
|
|
41 |
|
42 |
+
# SwissNER
|
43 |
+
|
44 |
+
A multilingual test set for named entity recognition (NER) on Swiss news articles.
|
45 |
+
|
46 |
+
## Description
|
47 |
+
|
48 |
+
SwissNER is a dataset for named entity recognition based on manually annotated news articles in Swiss Standard German, French, Italian, and Romansh Grischun.
|
49 |
+
We have manually annotated a selection of articles that have been published in February 2023 in the categories "Switzerland" or "Regional" on the following online news portals:
|
50 |
+
- Swiss Standard German: [srf.ch](https://www.srf.ch/)
|
51 |
+
- French: [rts.ch](https://www.rts.ch/)
|
52 |
+
- Italian: [rsi.ch](https://www.rsi.ch/)
|
53 |
+
- Romansh Grischun: [rtr.ch](https://www.rtr.ch/)
|
54 |
+
|
55 |
+
For each article we extracted the first two paragraphs after the lead paragraph.
|
56 |
+
We followed the guidelines of the CoNLL-2002 and 2003 shared tasks and annotated the names of persons, organizations, locations and miscellaneous entities.
|
57 |
+
The annotation was performed by a single annotator.
|
58 |
+
|
59 |
+
## License
|
60 |
+
- Text paragraphs: © Swiss Broadcasting Corporation (SRG SSR)
|
61 |
+
- Annotations: Attribution 4.0 International (CC BY 4.0)
|
62 |
+
|
63 |
+
## Statistics
|
64 |
+
| | DE | FR | IT | RM | Total |
|
65 |
+
|----------------------|-----:|------:|------:|------:|------:|
|
66 |
+
| Number of paragraphs | 200 | 200 | 200 | 200 | 800 |
|
67 |
+
| Number of tokens | 9498 | 11434 | 12423 | 13356 | 46711 |
|
68 |
+
| Number of entities | 479 | 475 | 556 | 591 | 2101 |
|
69 |
+
| – `PER` | 104 | 92 | 93 | 118 | 407 |
|
70 |
+
| – `ORG` | 193 | 216 | 266 | 227 | 902 |
|
71 |
+
| – `LOC` | 182 | 167 | 197 | 246 | 792 |
|
72 |
+
| – `MISC` | 113 | 79 | 88 | 39 | 319 |
|