kiddothe2b commited on
Commit
e13409a
1 Parent(s): c1144c7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +44 -12
README.md CHANGED
@@ -4,6 +4,8 @@ source_datasets:
4
  - extended
5
  language_creators:
6
  - found
 
 
7
  language:
8
  - bg
9
  - cs
@@ -42,13 +44,13 @@ EU Debates is a corpus of parliamentary proceedings (debates) from the EU parlia
42
  We exhaustively scrape the data from the official European Parliament Plenary website ([Link](https://www.europarl.europa.eu/)). All speeches are time-stamped, thematically organized on debates,
43
  and include metadata relevant to the speaker's identity (full name, euro-party affiliation, speaker role), and the debate (date and title).
44
  Older debate speeches are originally in English, while newer ones are linguistically diverse across the 23 official EU languages, thus we also provide machine-translated
45
- versions in English, when official translations are missing, using the EasyNMT framework with the M2M2-100 (418M) model.
46
 
47
  # Data Fields
48
 
49
- - `speaker_name: a `string` with the full-name of the speaker.
50
  - `speaker_party`: a `string` with the name of the euro-party (group) that the MEP is affiliated with.
51
- - `speaker_role`: a `string` with the role of speaker (Member of the European Parliament (MEP), EUROPARL President, etc.)
52
  - `debate_title`: a `string` with the title of the debate in the European Parliament.
53
  - `date`: a `string` with the full date (YYYY-MM-DD) of the speech.
54
  - `year` a `string` with the year (YYYY).
@@ -62,14 +64,14 @@ Example of a data instance from the EU Debates dataset:
62
 
63
  ```
64
  {
65
- 'speaker_name': 'Annemie Neyts-Uyttebroeck'
66
- 'speaker_party': 'ALDE',
67
- 'speaker_role': 'MEP', ,
68
- 'debate_title': 'Iran (debate)',
69
- 'date': '2009-07-15',
70
- 'year': '2009',
71
- 'text': 'Iran is a large country with a large and predominantly young population, a long and eventful history and an impressive culture. [...]'
72
- 'translated_text': None
73
  }
74
  ```
75
 
@@ -78,7 +80,7 @@ Example of a data instance from the EU Debates dataset:
78
  ```python
79
 
80
  from datasets import load_dataset
81
- eu_debates_dataset = load_dataset('coastalcph/eu_debates'))
82
 
83
  ```
84
 
@@ -121,6 +123,36 @@ Distribution of speeches across years and euro-parties:
121
  <tr><td> 2023 </td><td> 1716 </td><td> 1628 </td><td> 1040 </td><td> 878 </td><td> 619 </td><td> 779 </td><td> 795 </td><td> 499 </td><td> 7954 </td></tr>
122
  </table>
123
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
124
  ### Citation Information
125
 
126
  *[Llama meets EU: Investigating the European political spectrum through the lens of LLMs.
 
4
  - extended
5
  language_creators:
6
  - found
7
+ multilinguality:
8
+ - multilingual
9
  language:
10
  - bg
11
  - cs
 
44
  We exhaustively scrape the data from the official European Parliament Plenary website ([Link](https://www.europarl.europa.eu/)). All speeches are time-stamped, thematically organized on debates,
45
  and include metadata relevant to the speaker's identity (full name, euro-party affiliation, speaker role), and the debate (date and title).
46
  Older debate speeches are originally in English, while newer ones are linguistically diverse across the 23 official EU languages, thus we also provide machine-translated
47
+ versions in English, when official translations are missing, using the EasyNMT framework with the [M2M2-100 (418M)](https://huggingface.co/facebook/m2m100_418M) model (Fan et al., 2020).
48
 
49
  # Data Fields
50
 
51
+ - `speaker_name: a `string` with the full name of the speaker.
52
  - `speaker_party`: a `string` with the name of the euro-party (group) that the MEP is affiliated with.
53
+ - `speaker_role`: a `string` with the role of the speaker (Member of the European Parliament (MEP), EUROPARL President, etc.)
54
  - `debate_title`: a `string` with the title of the debate in the European Parliament.
55
  - `date`: a `string` with the full date (YYYY-MM-DD) of the speech.
56
  - `year` a `string` with the year (YYYY).
 
64
 
65
  ```
66
  {
67
+ 'speaker_name': 'Michèle Striffler'
68
+ 'speaker_party': 'PPE',
69
+ 'speaker_role': 'MEP',
70
+ 'debate_title': 'Famine in East Africa (debate)',
71
+ 'date': '2011-09-15',
72
+ 'year': '2011'
73
+ 'text': "Monsieur le Président, Madame le Commissaire, chers collègues, la situation humanitaire sans précédent que connaît la Corne de l'Afrique continue [...]",
74
+ 'translated_text': 'Mr. President, Mr. Commissioner, dear colleagues, the unprecedented humanitarian situation of the Horn of Africa continues [...]'}
75
  }
76
  ```
77
 
 
80
  ```python
81
 
82
  from datasets import load_dataset
83
+ eu_debates_dataset = load_dataset('coastalcph/eu_debates', split='train')
84
 
85
  ```
86
 
 
123
  <tr><td> 2023 </td><td> 1716 </td><td> 1628 </td><td> 1040 </td><td> 878 </td><td> 619 </td><td> 779 </td><td> 795 </td><td> 499 </td><td> 7954 </td></tr>
124
  </table>
125
 
126
+ Distribution of speeches across the 23 EU official languages:
127
+
128
+ | Language | Examples |
129
+ | ----------- | -------- |
130
+ | en | 40736 (46.7%) |
131
+ | de | 6497 (7.5%) |
132
+ | fr | 6024 (6.9%) |
133
+ | es | 5172 (5.9%) |
134
+ | it | 4506 (5.2%) |
135
+ | pl | 3792 (4.4%) |
136
+ | pt | 2713 (3.1%) |
137
+ | ro | 2308 (2.7%) |
138
+ | el | 2290 (2.6%) |
139
+ | nl | 2286 (2.6%) |
140
+ | hu | 1661 (1.9%) |
141
+ | hr | 1509 (1.7%) |
142
+ | cs | 1428 (1.6%) |
143
+ | sv | 1210 (1.4%) |
144
+ | bg | 928 (1.1%) |
145
+ | sk | 916 (1.1%) |
146
+ | sl | 753 (0.9%) |
147
+ | fi | 693 (0.8%) |
148
+ | lt | 618 (0.7%) |
149
+ | da | 578 (0.7%) |
150
+ | et | 342 (0.4%) |
151
+ | lv | 184 (0.2%) |
152
+ | mt | 0 (0.0%) |
153
+
154
+
155
+
156
  ### Citation Information
157
 
158
  *[Llama meets EU: Investigating the European political spectrum through the lens of LLMs.