odegiber commited on
Commit
e088039
1 Parent(s): dce351f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +64 -13
README.md CHANGED
@@ -14,7 +14,7 @@ size_categories:
14
  - unknown
15
  source_datasets: []
16
  task_categories:
17
- - conditional-text-generation
18
  task_ids:
19
  - summarization
20
  ---
@@ -32,12 +32,20 @@ task_ids:
32
  - [Data Fields](#data-fields)
33
  - [Data Splits](#data-splits)
34
  - [Dataset Creation](#dataset-creation)
 
 
 
 
 
 
 
 
35
  - [Additional Information](#additional-information)
36
  - [Dataset Curators](#dataset-curators)
37
  - [Licensing Information](#licensing-information)
38
  - [Citation Information](#citation-information)
39
- - [Funding](#funding)
40
-
41
  ## Dataset Description
42
 
43
  - **Paper:** [Sequence to Sequence Resources for Catalan](https://arxiv.org/pdf/2202.06871.pdf)
@@ -46,19 +54,18 @@ task_ids:
46
 
47
  ### Dataset Summary
48
 
49
- CaSum is a summarization dataset. It is extracted from a newswire corpus crawled from the Catalan News Agency. The corpus consists of 217,735 instances that are composed by the headline and the body.
50
 
51
  ### Supported Tasks and Leaderboards
52
 
53
- Summarization, Language Model
54
 
55
  ### Languages
56
 
57
- CA - Catalan
58
 
59
  ## Dataset Structure
60
 
61
-
62
  ### Data Instances
63
 
64
  ```
@@ -68,7 +75,6 @@ CA - Catalan
68
  }
69
  ```
70
 
71
-
72
  ### Data Fields
73
 
74
  - `summary` (str): Summary of the piece of news
@@ -84,16 +90,61 @@ We split our dataset into train, dev and test splits
84
 
85
  ## Dataset Creation
86
 
87
- We obtained each headline and its corresponding body and applied the following cleaning pipeline: deduplicating the documents, removing the documents with empty attributes, and deleting some boilerplate sentences.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
88
 
89
  ## Additional Information
 
90
  ### Dataset Curators
91
 
92
- Ona de Gibert, Jordi Armengol-Estapé and Maite Melero, from BSC-CNS, did the conversion and curation.
 
 
93
 
94
  ### Licensing information
95
 
96
- This work is licensed under a <a rel="license" href="https://creativecommons.org/licenses/by-nc/4.0/">Attribution-NonCommercial 4.0 International License</a>.
97
 
98
  ### BibTeX citation
99
 
@@ -110,6 +161,6 @@ If you use any of these resources (datasets or models) in your work, please cite
110
  }
111
  ```
112
 
113
- ### Funding
114
 
115
- This work was funded by the [MT4All CEF project](https://ec.europa.eu/inea/en/connecting-europe-facility/cef-telecom/2019-eu-ia-0031) and the [Catalan Ministry of the Vice-presidency, Digital Policies and Territory](https://politiquesdigitals.gencat.cat/en/inici/index.html) within the framework of the [Aina project](https://politiquesdigitals.gencat.cat/ca/tic/aina-el-projecte-per-garantir-el-catala-en-lera-digital/).
 
14
  - unknown
15
  source_datasets: []
16
  task_categories:
17
+ - summarization
18
  task_ids:
19
  - summarization
20
  ---
 
32
  - [Data Fields](#data-fields)
33
  - [Data Splits](#data-splits)
34
  - [Dataset Creation](#dataset-creation)
35
+ - [Curation Rationale](#curation-rationale)
36
+ - [Source Data](#source-data)
37
+ - [Annotations](#annotations)
38
+ - [Personal and Sensitive Information](#personal-and-sensitive-information)
39
+ - [Considerations for Using the Data](#considerations-for-using-the-data)
40
+ - [Social Impact of Dataset](#social-impact-of-dataset)
41
+ - [Discussion of Biases](#discussion-of-biases)
42
+ - [Other Known Limitations](#other-known-limitations)
43
  - [Additional Information](#additional-information)
44
  - [Dataset Curators](#dataset-curators)
45
  - [Licensing Information](#licensing-information)
46
  - [Citation Information](#citation-information)
47
+ - [Contributions](#contributions)
48
+
49
  ## Dataset Description
50
 
51
  - **Paper:** [Sequence to Sequence Resources for Catalan](https://arxiv.org/pdf/2202.06871.pdf)
 
54
 
55
  ### Dataset Summary
56
 
57
+ CaSum is a summarization dataset. It is extracted from a newswire corpus crawled from the [Catalan News Agency](https://www.acn.cat/). The corpus consists of 217,735 instances that are composed by the headline and the body.
58
 
59
  ### Supported Tasks and Leaderboards
60
 
61
+ - summarization: The dataset can be used to train a model for abstractive summarization. Success on this task is typically measured by achieving a high Rouge score. The [mbart-base-ca-casum](https://huggingface.co/projecte-aina/bart-base-ca-casum) model currently achieves a 41.39.
62
 
63
  ### Languages
64
 
65
+ The dataset is in Catalan (`ca-CA`).
66
 
67
  ## Dataset Structure
68
 
 
69
  ### Data Instances
70
 
71
  ```
 
75
  }
76
  ```
77
 
 
78
  ### Data Fields
79
 
80
  - `summary` (str): Summary of the piece of news
 
90
 
91
  ## Dataset Creation
92
 
93
+ ### Curation Rationale
94
+
95
+ We created this corpus to contribute to the development of language models in Catalan, a low-resource language. There exist few resources for summarization in Catalan.
96
+
97
+ ### Source Data
98
+
99
+ #### Initial Data Collection and Normalization
100
+
101
+ We obtained each headline and its corresponding body of each news piece on the [Catalan News Agency](https://www.acn.cat/) website and applied the following cleaning pipeline: deduplicating the documents, removing the documents with empty attributes, and deleting some boilerplate sentences.
102
+
103
+ #### Who are the source language producers?
104
+
105
+ The news portal [Catalan News Agency](https://www.acn.cat/).
106
+
107
+ ### Annotations
108
+
109
+ The dataset is unannotated.
110
+
111
+ #### Annotation process
112
+
113
+ [N/A]
114
+
115
+ #### Who are the annotators?
116
+
117
+ [N/A]
118
+
119
+ ### Personal and Sensitive Information
120
+
121
+ Since all data comes from public websites, no anonymization process was performed.
122
+
123
+ ## Considerations for Using the Data
124
+
125
+ ### Social Impact of Dataset
126
+
127
+ We hope this corpus contributes to the development of summarization models in Catalan, a low-resource language.
128
+
129
+ ### Discussion of Biases
130
+
131
+ We are aware that since the data comes from unreliable web pages, some biases may be present in the dataset. Nonetheless, we have not applied any steps to reduce their impact.
132
+
133
+ ### Other Known Limitations
134
+
135
+ [N/A]
136
 
137
  ## Additional Information
138
+
139
  ### Dataset Curators
140
 
141
+ Ona de Gibert Bonet, Barcelona Supercomputing Center (ona.degibert@bsc.es)
142
+
143
+ This work was funded by MT4All CEF project and the Departament de la Vicepresidència i de Polítiques Digitals i Territori de la Generalitat de Catalunya within the framework of the Aina project.
144
 
145
  ### Licensing information
146
 
147
+ [Creative Commons Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/).
148
 
149
  ### BibTeX citation
150
 
 
161
  }
162
  ```
163
 
164
+ ### Contributions
165
 
166
+ [N/A]