Datasets:
annotations_creators:
- found
language_creators:
- found
language:
- bg
- cs
- da
- de
- el
- en
- es
- et
- fi
- fr
- hr
- hu
- it
- lt
- lv
- mt
- nl
- pl
- pt
- ro
- sk
- sl
- sv
license:
- cc-by-sa-4.0
multilinguality:
- multilingual
size_categories:
- 10K<n<100K
source_datasets:
- original
task_categories:
- text-classification
task_ids:
- multi-label-classification
- topic-classification
pretty_name: MultiEURLEX
dataset_info:
- config_name: all_languages
features:
- name: celex_id
dtype: string
- name: text
dtype:
translation:
languages:
- en
- da
- de
- nl
- sv
- bg
- cs
- hr
- pl
- sk
- sl
- es
- fr
- it
- pt
- ro
- et
- fi
- hu
- lt
- lv
- el
- mt
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 6971500859
num_examples: 55000
- name: test
num_bytes: 1536038431
num_examples: 5000
- name: validation
num_bytes: 1062290624
num_examples: 5000
download_size: 2770050147
dataset_size: 9569829914
- config_name: bg
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 273160232
num_examples: 15986
- name: test
num_bytes: 109874757
num_examples: 5000
- name: validation
num_bytes: 76892269
num_examples: 5000
download_size: 164279607
dataset_size: 459927258
- config_name: cs
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 189826374
num_examples: 23187
- name: test
num_bytes: 60702802
num_examples: 5000
- name: validation
num_bytes: 42764231
num_examples: 5000
download_size: 132411256
dataset_size: 293293407
- config_name: da
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 395774705
num_examples: 55000
- name: test
num_bytes: 60343684
num_examples: 5000
- name: validation
num_bytes: 42366378
num_examples: 5000
download_size: 215874886
dataset_size: 498484767
- config_name: de
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 425489833
num_examples: 55000
- name: test
num_bytes: 65739062
num_examples: 5000
- name: validation
num_bytes: 46079562
num_examples: 5000
download_size: 232089961
dataset_size: 537308457
- config_name: el
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 768224743
num_examples: 55000
- name: test
num_bytes: 117209312
num_examples: 5000
- name: validation
num_bytes: 81923366
num_examples: 5000
download_size: 2770050147
dataset_size: 967357421
- config_name: en
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 389250111
num_examples: 55000
- name: test
num_bytes: 58966951
num_examples: 5000
- name: validation
num_bytes: 41516153
num_examples: 5000
download_size: 206930941
dataset_size: 489733215
- config_name: es
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 433955311
num_examples: 52785
- name: test
num_bytes: 66884992
num_examples: 5000
- name: validation
num_bytes: 47178809
num_examples: 5000
download_size: 231656657
dataset_size: 548019112
- config_name: et
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 173878703
num_examples: 23126
- name: test
num_bytes: 56535287
num_examples: 5000
- name: validation
num_bytes: 39580866
num_examples: 5000
download_size: 2770050147
dataset_size: 269994856
- config_name: fi
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 336145949
num_examples: 42497
- name: test
num_bytes: 63280920
num_examples: 5000
- name: validation
num_bytes: 44500040
num_examples: 5000
download_size: 2770050147
dataset_size: 443926909
- config_name: fr
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 442358833
num_examples: 55000
- name: test
num_bytes: 68520115
num_examples: 5000
- name: validation
num_bytes: 48408926
num_examples: 5000
download_size: 238412621
dataset_size: 559287874
- config_name: hr
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 80808161
num_examples: 7944
- name: test
num_bytes: 56790818
num_examples: 5000
- name: validation
num_bytes: 23881820
num_examples: 2500
download_size: 75125923
dataset_size: 161480799
- config_name: hu
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 208805862
num_examples: 22664
- name: test
num_bytes: 68990666
num_examples: 5000
- name: validation
num_bytes: 48101023
num_examples: 5000
download_size: 2770050147
dataset_size: 325897551
- config_name: it
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 429495741
num_examples: 55000
- name: test
num_bytes: 64731758
num_examples: 5000
- name: validation
num_bytes: 45886525
num_examples: 5000
download_size: 234661012
dataset_size: 540114024
- config_name: lt
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 185211691
num_examples: 23188
- name: test
num_bytes: 59484711
num_examples: 5000
- name: validation
num_bytes: 41841024
num_examples: 5000
download_size: 2770050147
dataset_size: 286537426
- config_name: lv
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 186396252
num_examples: 23208
- name: test
num_bytes: 59814093
num_examples: 5000
- name: validation
num_bytes: 42002727
num_examples: 5000
download_size: 2770050147
dataset_size: 288213072
- config_name: mt
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 179866781
num_examples: 17521
- name: test
num_bytes: 65831230
num_examples: 5000
- name: validation
num_bytes: 46737914
num_examples: 5000
download_size: 2770050147
dataset_size: 292435925
- config_name: nl
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 430232711
num_examples: 55000
- name: test
num_bytes: 64728022
num_examples: 5000
- name: validation
num_bytes: 45452538
num_examples: 5000
download_size: 230199167
dataset_size: 540413271
- config_name: pl
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 202211442
num_examples: 23197
- name: test
num_bytes: 64654967
num_examples: 5000
- name: validation
num_bytes: 45545505
num_examples: 5000
download_size: 139058173
dataset_size: 312411914
- config_name: pt
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 419281855
num_examples: 52370
- name: test
num_bytes: 64771235
num_examples: 5000
- name: validation
num_bytes: 45897219
num_examples: 5000
download_size: 227524717
dataset_size: 529950309
- config_name: ro
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 164966676
num_examples: 15921
- name: test
num_bytes: 67248472
num_examples: 5000
- name: validation
num_bytes: 46968070
num_examples: 5000
download_size: 2770050147
dataset_size: 279183218
- config_name: sk
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 188126733
num_examples: 22971
- name: test
num_bytes: 60922674
num_examples: 5000
- name: validation
num_bytes: 42786781
num_examples: 5000
download_size: 134875274
dataset_size: 291836188
- config_name: sl
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 170800897
num_examples: 23184
- name: test
num_bytes: 54552429
num_examples: 5000
- name: validation
num_bytes: 38286410
num_examples: 5000
download_size: 124008905
dataset_size: 263639736
- config_name: sv
features:
- name: celex_id
dtype: string
- name: text
dtype: string
- name: labels
sequence:
class_label:
names:
'0': '100149'
'1': '100160'
'2': '100148'
'3': '100147'
'4': '100152'
'5': '100143'
'6': '100156'
'7': '100158'
'8': '100154'
'9': '100153'
'10': '100142'
'11': '100145'
'12': '100150'
'13': '100162'
'14': '100159'
'15': '100144'
'16': '100151'
'17': '100157'
'18': '100161'
'19': '100146'
'20': '100155'
splits:
- name: train
num_bytes: 329071237
num_examples: 42490
- name: test
num_bytes: 60602014
num_examples: 5000
- name: validation
num_bytes: 42766055
num_examples: 5000
download_size: 184902175
dataset_size: 432439306
configs:
- config_name: bg
data_files:
- split: train
path: bg/train-*
- split: test
path: bg/test-*
- split: validation
path: bg/validation-*
- config_name: cs
data_files:
- split: train
path: cs/train-*
- split: test
path: cs/test-*
- split: validation
path: cs/validation-*
- config_name: da
data_files:
- split: train
path: da/train-*
- split: test
path: da/test-*
- split: validation
path: da/validation-*
- config_name: de
data_files:
- split: train
path: de/train-*
- split: test
path: de/test-*
- split: validation
path: de/validation-*
- config_name: en
data_files:
- split: train
path: en/train-*
- split: test
path: en/test-*
- split: validation
path: en/validation-*
- config_name: es
data_files:
- split: train
path: es/train-*
- split: test
path: es/test-*
- split: validation
path: es/validation-*
- config_name: fr
data_files:
- split: train
path: fr/train-*
- split: test
path: fr/test-*
- split: validation
path: fr/validation-*
- config_name: hr
data_files:
- split: train
path: hr/train-*
- split: test
path: hr/test-*
- split: validation
path: hr/validation-*
- config_name: it
data_files:
- split: train
path: it/train-*
- split: test
path: it/test-*
- split: validation
path: it/validation-*
- config_name: nl
data_files:
- split: train
path: nl/train-*
- split: test
path: nl/test-*
- split: validation
path: nl/validation-*
- config_name: pl
data_files:
- split: train
path: pl/train-*
- split: test
path: pl/test-*
- split: validation
path: pl/validation-*
- config_name: pt
data_files:
- split: train
path: pt/train-*
- split: test
path: pt/test-*
- split: validation
path: pt/validation-*
- config_name: sk
data_files:
- split: train
path: sk/train-*
- split: test
path: sk/test-*
- split: validation
path: sk/validation-*
- config_name: sl
data_files:
- split: train
path: sl/train-*
- split: test
path: sl/test-*
- split: validation
path: sl/validation-*
- config_name: sv
data_files:
- split: train
path: sv/train-*
- split: test
path: sv/test-*
- split: validation
path: sv/validation-*
Dataset Card for "MultiEURLEX"
Table of Contents
- Dataset Description
- Dataset Structure
- Dataset Creation
- Considerations for Using the Data
- Additional Information
Dataset Description
- Repository: https://github.com/nlpaueb/multi-eurlex
- Paper: https://arxiv.org/abs/2109.00904
- Data: https://doi.org/10.5281/zenodo.5363165
- Leaderboard: N/A
- Point of Contact: Ilias Chalkidis
Dataset Summary
Documents
MultiEURLEX 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.
Multi-granular Labeling
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. We created three alternative sets of labels per document, by replacing each assigned concept by its ancestor from level 1, 2, or 3, respectively. 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.
Data Split and Concept Drift
MultiEURLEX is chronologically split in training (55k, 1958-2010), development (5k, 2010-2012), test (5k, 2012-2016) subsets, using the English documents. The test subset contains the same 5k documents in all 23 languages. The development subset also contains the same 5k documents in 23 languages, except Croatian. Croatia is the most recent EU member (2013); older laws are gradually translated. For the official languages of the seven oldest member countries, the same 55k training documents are available; for the other languages, only a subset of the 55k training documents is available. Compared to EUR-LEX (Chalkidis et al., 2019), MultiEURLEX is not only larger (8k more documents) and multilingual; it is also more challenging, as the chronological split leads to temporal real-world concept drift across the training, development, test subsets, i.e., differences in label distribution and phrasing, representing a realistic temporal generalization problem (Huang et al., 2019; Lazaridou et al., 2021). Recently, S酶gaard et al. (2021) showed this setup is more realistic, as it does not over-estimate real performance, contrary to random splits (Gorman and Bedrick, 2019).
Supported Tasks and Leaderboards
Similarly to EUR-LEX (Chalkidis et al., 2019), MultiEURLEX can be used for legal topic classification, a multi-label classification task where legal documents need to be assigned concepts (in our case, from EUROVOC) reflecting their topics. Unlike EUR-LEX, however, MultiEURLEX supports labels from three different granularities (EUROVOC levels). More importantly, apart from monolingual (one-to-one) experiments, it can be used to study cross-lingual transfer scenarios, including one-to-many (systems trained in one language and used in other languages with no training data), and many-to-one or many-to-many (systems jointly trained in multiple languages and used in one or more other languages).
The dataset is not yet part of an established benchmark.
Languages
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.
Dataset Structure
Data Instances
Multilingual use of the dataset
When the dataset is used in a multilingual setting selecting the the 'all_languages' flag:
from datasets import load_dataset
dataset = load_dataset('multi_eurlex', 'all_languages')
{
"celex_id": "31979D0509",
"text": {"en": "COUNCIL DECISION of 24 May 1979 on financial aid from the Community for the eradication of African swine fever in Spain (79/509/EEC)\nTHE COUNCIL OF THE EUROPEAN COMMUNITIES\nHaving regard to the Treaty establishing the European Economic Community, and in particular Article 43 thereof,\nHaving regard to the proposal from the Commission (1),\nHaving regard to the opinion of the European Parliament (2),\nWhereas the Community should take all appropriate measures to protect itself against the appearance of African swine fever on its territory;\nWhereas to this end the Community has undertaken, and continues to undertake, action designed to contain outbreaks of this type of disease far from its frontiers by helping countries affected to reinforce their preventive measures ; whereas for this purpose Community subsidies have already been granted to Spain;\nWhereas these measures have unquestionably made an effective contribution to the protection of Community livestock, especially through the creation and maintenance of a buffer zone north of the river Ebro;\nWhereas, however, in the opinion of the Spanish authorities themselves, the measures so far implemented must be reinforced if the fundamental objective of eradicating the disease from the entire country is to be achieved;\nWhereas the Spanish authorities have asked the Community to contribute to the expenses necessary for the efficient implementation of a total eradication programme;\nWhereas a favourable response should be given to this request by granting aid to Spain, having regard to the undertaking given by that country to protect the Community against African swine fever and to eliminate completely this disease by the end of a five-year eradication plan;\nWhereas this eradication plan must include certain measures which guarantee the effectiveness of the action taken, and it must be possible to adapt these measures to developments in the situation by means of a procedure establishing close cooperation between the Member States and the Commission;\nWhereas it is necessary to keep the Member States regularly informed as to the progress of the action undertaken,",
"es": "DECISI脫N DEL CONSEJO de 24 de mayo de 1979 sobre ayuda financiera de la Comunidad para la erradicaci贸n de la peste porcina africana en Espa帽a (79/509/CEE)\nEL CONSEJO DE LAS COMUNIDADES EUROPEAS\nVeniendo en cuenta el Tratado constitutivo de la Comunidad Econ贸mica Europea y, en particular, Su art铆culo 43,\n Vista la propuesta de la Comisi贸n (1),\n Visto el dictamen del Parlamento Europeo (2),\nConsiderando que la Comunidad debe tomar todas las medidas adecuadas para protegerse contra la aparici贸n de la peste porcina africana en su territorio;\nConsiderando a tal fin que la Comunidad ha emprendido y sigue llevando a cabo acciones destinadas a contener los brotes de este tipo de enfermedades lejos de sus fronteras, ayudando a los pa铆ses afectados a reforzar sus medidas preventivas; que a tal efecto ya se han concedido a Espa帽a subvenciones comunitarias;\nQue estas medidas han contribuido sin duda alguna a la protecci贸n de la ganader铆a comunitaria, especialmente mediante la creaci贸n y mantenimiento de una zona tamp贸n al norte del r铆o Ebro;\nConsiderando, no obstante, , a juicio de las propias autoridades espa帽olas, las medidas implementadas hasta ahora deben reforzarse si se quiere alcanzar el objetivo fundamental de erradicar la enfermedad en todo el pa铆s;\nConsiderando que las autoridades espa帽olas han pedido a la Comunidad que contribuya a los gastos necesarios para la ejecuci贸n eficaz de un programa de erradicaci贸n total;\nConsiderando que conviene dar una respuesta favorable a esta solicitud concediendo una ayuda a Espa帽a, habida cuenta del compromiso asumido por dicho pa铆s de proteger a la Comunidad contra la peste porcina africana y de eliminar completamente esta enfermedad al final de un plan de erradicaci贸n de cinco a帽os;\nMientras que este plan de erradicaci贸n debe incluir e determinadas medidas que garanticen la eficacia de las acciones emprendidas, debiendo ser posible adaptar estas medidas a la evoluci贸n de la situaci贸n mediante un procedimiento que establezca una estrecha cooperaci贸n entre los Estados miembros y la Comisi贸n;\nConsiderando que es necesario mantener el Los Estados miembros informados peri贸dicamente sobre el progreso de las acciones emprendidas.",
"de": "...",
"bg": "..."
},
"labels": [
1,
13,
47
]
}
Monolingual use of the dataset
When the dataset is used in a monolingual setting selecting the ISO language code for one of the 23 supported languages. For example:
from datasets import load_dataset
dataset = load_dataset('multi_eurlex', 'en')
{
"celex_id": "31979D0509",
"text": "COUNCIL DECISION of 24 May 1979 on financial aid from the Community for the eradication of African swine fever in Spain (79/509/EEC)\nTHE COUNCIL OF THE EUROPEAN COMMUNITIES\nHaving regard to the Treaty establishing the European Economic Community, and in particular Article 43 thereof,\nHaving regard to the proposal from the Commission (1),\nHaving regard to the opinion of the European Parliament (2),\nWhereas the Community should take all appropriate measures to protect itself against the appearance of African swine fever on its territory;\nWhereas to this end the Community has undertaken, and continues to undertake, action designed to contain outbreaks of this type of disease far from its frontiers by helping countries affected to reinforce their preventive measures ; whereas for this purpose Community subsidies have already been granted to Spain;\nWhereas these measures have unquestionably made an effective contribution to the protection of Community livestock, especially through the creation and maintenance of a buffer zone north of the river Ebro;\nWhereas, however, in the opinion of the Spanish authorities themselves, the measures so far implemented must be reinforced if the fundamental objective of eradicating the disease from the entire country is to be achieved;\nWhereas the Spanish authorities have asked the Community to contribute to the expenses necessary for the efficient implementation of a total eradication programme;\nWhereas a favourable response should be given to this request by granting aid to Spain, having regard to the undertaking given by that country to protect the Community against African swine fever and to eliminate completely this disease by the end of a five-year eradication plan;\nWhereas this eradication plan must include certain measures which guarantee the effectiveness of the action taken, and it must be possible to adapt these measures to developments in the situation by means of a procedure establishing close cooperation between the Member States and the Commission;\nWhereas it is necessary to keep the Member States regularly informed as to the progress of the action undertaken,",
"labels": [
1,
13,
47
]
}
Data Fields
Multilingual use of the dataset
The following data fields are provided for documents (train, dev, test):
celex_id: (str) The official ID of the document. The CELEX number is the unique identifier for all publications in both Eur-Lex and CELLAR.text: (dict[str]) A dictionary with the 23 languages as keys and the full content of each document as values.labels: (List[int]) The relevant EUROVOC concepts (labels).
Monolingual use of the dataset
The following data fields are provided for documents (train, dev, test):
celex_id: (str) The official ID of the document. The CELEX number is the unique identifier for all publications in both Eur-Lex and CELLAR.text: (str) The full content of each document across languages.labels: (List[int]) The relevant EUROVOC concepts (labels).
If you want to use the descriptors of the EUROVOC concepts, similar to Chalkidis et al. (2020), please download the relevant JSON file here. Then you may load it and use it:
import json
from datasets import load_dataset
# Load the English part of the dataset
dataset = load_dataset('multi_eurlex', 'en', split='train')
# Load (label_id, descriptor) mapping
with open('./eurovoc_descriptors.json') as jsonl_file:
eurovoc_concepts = json.load(jsonl_file)
# Get feature map info
classlabel = dataset.features["labels"].feature
# Retrieve IDs and descriptors from dataset
for sample in dataset:
print(f'DOCUMENT: {sample["celex_id"]}')
# DOCUMENT: 32006D0213
for label_id in sample['labels']:
print(f'LABEL: id:{label_id}, eurovoc_id: {classlabel.int2str(label_id)}, \
eurovoc_desc:{eurovoc_concepts[classlabel.int2str(label_id)]}')
# LABEL: id: 1, eurovoc_id: '100160', eurovoc_desc: 'industry'
Data Splits
| Language | ISO code | Member Countries where official | EU Speakers [1] | Number of Documents [2] |
| English | en | United Kingdom (1973-2020), Ireland (1973), Malta (2004) | 13/ 51% | 55,000 / 5,000 / 5,000 |
| German | de | Germany (1958), Belgium (1958), Luxembourg (1958) | 16/32% | 55,000 / 5,000 / 5,000 |
| French | fr | France (1958), Belgium(1958), Luxembourg (1958) | 12/26% | 55,000 / 5,000 / 5,000 |
| Italian | it | Italy (1958) | 13/16% | 55,000 / 5,000 / 5,000 |
| Spanish | es | Spain (1986) | 8/15% | 52,785 / 5,000 / 5,000 |
| Polish | pl | Poland (2004) | 8/9% | 23,197 / 5,000 / 5,000 |
| Romanian | ro | Romania (2007) | 5/5% | 15,921 / 5,000 / 5,000 |
| Dutch | nl | Netherlands (1958), Belgium (1958) | 4/5% | 55,000 / 5,000 / 5,000 |
| Greek | el | Greece (1981), Cyprus (2008) | 3/4% | 55,000 / 5,000 / 5,000 |
| Hungarian | hu | Hungary (2004) | 3/3% | 22,664 / 5,000 / 5,000 |
| Portuguese | pt | Portugal (1986) | 2/3% | 23,188 / 5,000 / 5,000 |
| Czech | cs | Czech Republic (2004) | 2/3% | 23,187 / 5,000 / 5,000 |
| Swedish | sv | Sweden (1995) | 2/3% | 42,490 / 5,000 / 5,000 |
| Bulgarian | bg | Bulgaria (2007) | 2/2% | 15,986 / 5,000 / 5,000 |
| Danish | da | Denmark (1973) | 1/1% | 55,000 / 5,000 / 5,000 |
| Finnish | fi | Finland (1995) | 1/1% | 42,497 / 5,000 / 5,000 |
| Slovak | sk | Slovakia (2004) | 1/1% | 15,986 / 5,000 / 5,000 |
| Lithuanian | lt | Lithuania (2004) | 1/1% | 23,188 / 5,000 / 5,000 |
| Croatian | hr | Croatia (2013) | 1/1% | 7,944 / 2,500 / 5,000 |
| Slovene | sl | Slovenia (2004) | <1/<1% | 23,184 / 5,000 / 5,000 |
| Estonian | et | Estonia (2004) | <1/<1% | 23,126 / 5,000 / 5,000 |
| Latvian | lv | Latvia (2004) | <1/<1% | 23,188 / 5,000 / 5,000 |
| Maltese | mt | Malta (2004) | <1/<1% | 17,521 / 5,000 / 5,000 |
[1] Native and Total EU speakers percentage (%)
[2] Training / Development / Test Splits
Dataset Creation
Curation Rationale
The dataset was curated by Chalkidis et al. (2021).
The documents have been annotated by the Publications Office of EU (https://publications.europa.eu/en).
Source Data
Initial Data Collection and Normalization
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). We stripped HTML mark-up to provide the documents in plain text format. 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.
Who are the source language producers?
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.
Annotations
Annotation process
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. 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. 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.
Who are the annotators?
Publications Office of EU (https://publications.europa.eu/en)
Personal and Sensitive Information
The dataset contains publicly available EU laws that do not include personal or sensitive information with the exception of trivial information presented by consent, e.g., the names of the current presidents of the European Parliament and European Council, and other administration bodies.
Considerations for Using the Data
Social Impact of Dataset
[More Information Needed]
Discussion of Biases
[More Information Needed]
Other Known Limitations
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).
Additional Information
Dataset Curators
Chalkidis et al. (2021)
Licensing Information
We provide MultiEURLEX with the same licensing as the original EU data (CC-BY-4.0):
漏 European Union, 1998-2021
The Commission鈥檚 document reuse policy is based on Decision 2011/833/EU. Unless otherwise specified, you can re-use the legal documents published in EUR-Lex for commercial or non-commercial purposes.
The copyright for the editorial content of this website, the summaries of EU legislation and the consolidated texts, which is owned by the EU, is licensed under the Creative Commons Attribution 4.0 International licence. This means that you can re-use the content provided you acknowledge the source and indicate any changes you have made.
Source: https://eur-lex.europa.eu/content/legal-notice/legal-notice.html
Read more: https://eur-lex.europa.eu/content/help/faq/reuse-contents-eurlex.html
Citation Information
Ilias Chalkidis, Manos Fergadiotis, and Ion Androutsopoulos. MultiEURLEX - A multi-lingual and multi-label legal document classification dataset for zero-shot cross-lingual transfer. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. Punta Cana, Dominican Republic. 2021
@InProceedings{chalkidis-etal-2021-multieurlex,
author = {Chalkidis, Ilias
and Fergadiotis, Manos
and Androutsopoulos, Ion},
title = {MultiEURLEX -- A multi-lingual and multi-label legal document
classification dataset for zero-shot cross-lingual transfer},
booktitle = {Proceedings of the 2021 Conference on Empirical Methods
in Natural Language Processing},
year = {2021},
publisher = {Association for Computational Linguistics},
location = {Punta Cana, Dominican Republic},
url = {https://arxiv.org/abs/2109.00904}
}
Contributions
Thanks to @iliaschalkidis for adding this dataset.