nlpaueb commited on
Commit
eb324fc
1 Parent(s): 692edd7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -14
README.md CHANGED
@@ -66,7 +66,7 @@ task_ids:
66
 
67
  MultiEURLEX of Chalkidis et al. (2021) comprises 65k EU laws in 23 official EU languages. Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU. Each EUROVOC label ID is associated with a *label descriptor*, e.g., [60, agri-foodstuffs], [6006, plant product], [1115, fruit]. The descriptors are also available in the 23 languages. Chalkidis et al. (2019) published a monolingual (English) version of this dataset, called EUR-LEX, comprising 57k EU laws with the originally assigned gold labels.
68
 
69
- In this new version, dubbed Non-Parallel MultiEURLEX (incl. Translations), MultiEURLEX comprises non-parallel documents across 5 languages (English, German, French, Greek, and Slovak), i.e., 11,000 different documents per language, including also translations from English to the rest of the 4 available languages.
70
 
71
  ### Supported Tasks and Leaderboards
72
 
@@ -78,7 +78,7 @@ The dataset is not yet part of an established benchmark.
78
 
79
  The EU has 24 official languages. When new members join the EU, the set of official languages usually expands, except the languages are already included. MultiEURLEX covers 23 languages from seven language families (Germanic, Romance, Slavic, Uralic, Baltic, Semitic, Hellenic). EU laws are published in all official languages, except Irish, for resource-related reasons (Read more at https://europa.eu/european-union/about-eu/eu-languages_en). This wide coverage makes MultiEURLEX a valuable testbed for cross-lingual transfer. All languages use the Latin script, except for Bulgarian (Cyrillic script) and Greek. Several other languages are also spoken in EU countries. The EU is home to over 60 additional indigenous regional or minority languages, e.g., Basque, Catalan, Frisian, Saami, and Yiddish, among others, spoken by approx. 40 million people, but these additional languages are not considered official (in terms of EU), and EU laws are not translated to them.
80
 
81
- This version of MultiEURLEX covers 5 EU languages (English, German, French, Greek, and Slovak).
82
 
83
  ## Dataset Structure
84
 
@@ -100,7 +100,7 @@ dataset = load_dataset('nlpaueb/multi_eurlex', 'all_languages')
100
  "en2fr": "DU CONSEIL du 24 mai 1979 concernant l'aide financiere de la Communaute e l'eradication de la peste porcine africaine en Espagne (79/509/CEE)\nLE CONSEIL DES COMMUNAUTAS EUROPENNES ...",
101
  "en2de": "...",
102
  "en2el": "...",
103
- "en2el": "..."
104
  },
105
  "labels": [
106
  1,
@@ -112,7 +112,7 @@ dataset = load_dataset('nlpaueb/multi_eurlex', 'all_languages')
112
 
113
  **Monolingual use of the dataset**
114
 
115
- When the dataset is used in a monolingual setting selecting the ISO language code for one of the 23 supported languages. For example:
116
 
117
  ```python
118
  from datasets import load_dataset
@@ -203,21 +203,19 @@ The documents have been annotated by the Publications Office of EU (https://publ
203
  #### Initial Data Collection and Normalization
204
 
205
  The original data are available at the EUR-LEX portal (https://eur-lex.europa.eu) in unprocessed formats (HTML, XML, RDF). The documents were downloaded from the EUR-LEX portal in HTML. The relevant EUROVOC concepts were downloaded from the SPARQL endpoint of the Publications Office of EU (http://publications.europa.eu/webapi/rdf/sparql).
206
- We stripped HTML mark-up to provide the documents in plain text format.
207
- We inferred the labels for EUROVOC levels 1--3, by backtracking the EUROVOC hierarchy branches, from the originally assigned labels to their ancestors in levels 1--3, respectively.
208
 
209
  #### Who are the source language producers?
210
 
211
- The EU has 24 official languages. When new members join the EU, the set of official languages usually expands, except the languages are already included. MultiEURLEX covers 23 languages from seven language families (Germanic, Romance, Slavic, Uralic, Baltic, Semitic, Hellenic). EU laws are published in all official languages, except Irish, for resource-related reasons (Read more at https://europa.eu/european-union/about-eu/eu-languages_en). This wide coverage makes MultiEURLEX a valuable testbed for cross-lingual transfer. All languages use the Latin script, except for Bulgarian (Cyrillic script) and Greek. Several other languages are also spoken in EU countries. The EU is home to over 60 additional indigenous regional or minority languages, e.g., Basque, Catalan, Frisian, Saami, and Yiddish, among others, spoken by approx. 40 million people, but these additional languages are not considered official (in terms of EU), and EU laws are not translated to them.
212
-
213
 
214
  ### Annotations
215
 
216
  #### Annotation process
217
 
218
  All the documents of the dataset have been annotated by the Publications Office of EU (https://publications.europa.eu/en) with multiple concepts from EUROVOC (http://eurovoc.europa.eu/). EUROVOC has eight levels of concepts. Each document is assigned one or more concepts (labels). If a document is assigned a concept, the ancestors and descendants of that concept are typically not assigned to the same document. The documents were originally annotated with concepts from levels 3 to 8.
219
- We augmented the annotation with three alternative sets of labels per document, replacing each assigned concept by its ancestor from level 1, 2, or 3, respectively.
220
- Thus, we provide four sets of gold labels per document, one for each of the first three levels of the hierarchy, plus the original sparse label assignment.Levels 4 to 8 cannot be used independently, as many documents have gold concepts from the third level; thus many documents will be mislabeled, if we discard level 3.
221
 
222
 
223
  #### Who are the annotators?
@@ -240,7 +238,7 @@ The dataset contains publicly available EU laws that do not include personal or
240
 
241
  ### Other Known Limitations
242
 
243
- MultiEURLEX covers 23 languages from seven language families (Germanic, Romance, Slavic, Uralic, Baltic, Semitic, Hellenic). This does not imply that no other languages are spoken in EU countries, although EU laws are not translated to other languages (https://europa.eu/european-union/about-eu/eu-languages_en).
244
 
245
  ## Additional Information
246
 
@@ -263,9 +261,9 @@ Read more: https://eur-lex.europa.eu/content/help/faq/reuse-contents-eurlex.htm
263
 
264
  ### Citation Information
265
 
266
- *Ilias Chalkidis, Manos Fergadiotis, and Ion Androutsopoulos.*
267
- *MultiEURLEX - A multi-lingual and multi-label legal document classification dataset for zero-shot cross-lingual transfer.*
268
- *Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. Punta Cana, Dominican Republic. 2021*
269
  ```
270
  @InProceedings{xenouleas-etal-2022-realistic-multieurlex,
271
  author = {Xenouleas, Stratos
 
66
 
67
  MultiEURLEX of Chalkidis et al. (2021) comprises 65k EU laws in 23 official EU languages. Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU. Each EUROVOC label ID is associated with a *label descriptor*, e.g., [60, agri-foodstuffs], [6006, plant product], [1115, fruit]. The descriptors are also available in the 23 languages. Chalkidis et al. (2019) published a monolingual (English) version of this dataset, called EUR-LEX, comprising 57k EU laws with the originally assigned gold labels.
68
 
69
+ In this new version, dubbed "Non-Parallel MultiEURLEX (incl. Translations)", MultiEURLEX comprises non-parallel documents across 5 languages (English, German, French, Greek, and Slovak), i.e., 11,000 different documents per language, including also translations from English to the rest of the 4 available languages.
70
 
71
  ### Supported Tasks and Leaderboards
72
 
 
78
 
79
  The EU has 24 official languages. When new members join the EU, the set of official languages usually expands, except the languages are already included. MultiEURLEX covers 23 languages from seven language families (Germanic, Romance, Slavic, Uralic, Baltic, Semitic, Hellenic). EU laws are published in all official languages, except Irish, for resource-related reasons (Read more at https://europa.eu/european-union/about-eu/eu-languages_en). This wide coverage makes MultiEURLEX a valuable testbed for cross-lingual transfer. All languages use the Latin script, except for Bulgarian (Cyrillic script) and Greek. Several other languages are also spoken in EU countries. The EU is home to over 60 additional indigenous regional or minority languages, e.g., Basque, Catalan, Frisian, Saami, and Yiddish, among others, spoken by approx. 40 million people, but these additional languages are not considered official (in terms of EU), and EU laws are not translated to them.
80
 
81
+ This version of MultiEURLEX covers 5 EU languages (English, German, French, Greek, and Slovak). It also includes machine-translated versions of the documents using the EasyNMT framework (https://github.com/UKPLab/EasyNMT) utilizing the many-to-many M2M_100_418M model of Fan et al. (2020) for el-to-en and el-to-de pairs and the OPUS-MT (Tiedemann et al., 2020) models for the rest.
82
 
83
  ## Dataset Structure
84
 
 
100
  "en2fr": "DU CONSEIL du 24 mai 1979 concernant l'aide financiere de la Communaute e l'eradication de la peste porcine africaine en Espagne (79/509/CEE)\nLE CONSEIL DES COMMUNAUTAS EUROPENNES ...",
101
  "en2de": "...",
102
  "en2el": "...",
103
+ "en2sk": "..."
104
  },
105
  "labels": [
106
  1,
 
112
 
113
  **Monolingual use of the dataset**
114
 
115
+ When the dataset is used in a monolingual setting selecting the ISO language code for one of the 5 supported languages, or supported translation pairs in the form src2trg, where src and trg are ISO language codes, e.g., en2fr for English translated to French. For example:
116
 
117
  ```python
118
  from datasets import load_dataset
 
203
  #### Initial Data Collection and Normalization
204
 
205
  The original data are available at the EUR-LEX portal (https://eur-lex.europa.eu) in unprocessed formats (HTML, XML, RDF). The documents were downloaded from the EUR-LEX portal in HTML. The relevant EUROVOC concepts were downloaded from the SPARQL endpoint of the Publications Office of EU (http://publications.europa.eu/webapi/rdf/sparql).
206
+ Chalkidis et al. (2021) stripped HTML mark-up to provide the documents in plain text format and inferred the labels for EUROVOC levels 1--3, by backtracking the EUROVOC hierarchy branches, from the originally assigned labels to their ancestors in levels 1--3, respectively.
 
207
 
208
  #### Who are the source language producers?
209
 
210
+ [More Information Needed]
 
211
 
212
  ### Annotations
213
 
214
  #### Annotation process
215
 
216
  All the documents of the dataset have been annotated by the Publications Office of EU (https://publications.europa.eu/en) with multiple concepts from EUROVOC (http://eurovoc.europa.eu/). EUROVOC has eight levels of concepts. Each document is assigned one or more concepts (labels). If a document is assigned a concept, the ancestors and descendants of that concept are typically not assigned to the same document. The documents were originally annotated with concepts from levels 3 to 8.
217
+ Chalkidis et al. (2021)augmented the annotation with three alternative sets of labels per document, replacing each assigned concept by its ancestor from level 1, 2, or 3, respectively.
218
+ Thus, Chalkidis et al. (2021) provide four sets of gold labels per document, one for each of the first three levels of the hierarchy, plus the original sparse label assignment.Levels 4 to 8 cannot be used independently, as many documents have gold concepts from the third level; thus many documents will be mislabeled, if we discard level 3.
219
 
220
 
221
  #### Who are the annotators?
 
238
 
239
  ### Other Known Limitations
240
 
241
+ [More Information Needed]
242
 
243
  ## Additional Information
244
 
 
261
 
262
  ### Citation Information
263
 
264
+ *Stratos Xenouleas, Alexi Tsoukara, Giannis Panagiotakis Ilias Chalkidis, and Ion Androutsopoulos.*
265
+ *Realistic Zero-Shot Cross-Lingual Transfer in Legal Topic Classification.*
266
+ *Proceedings of 12th Hellenic Conference on Artificial Intelligence (SETN 2022). Corfu, Greece. 2022*
267
  ```
268
  @InProceedings{xenouleas-etal-2022-realistic-multieurlex,
269
  author = {Xenouleas, Stratos