Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -78,19 +78,31 @@ This dataset is in Catalan (ca-ES). Coming from the web, some documents may cont
|
|
78 |
|
79 |
### Data Instances
|
80 |
|
81 |
-
The dataset is provided in a
|
82 |
|
83 |
```
|
84 |
-
|
85 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
86 |
```
|
87 |
|
88 |
### Data Fields
|
89 |
|
90 |
-
- `
|
91 |
- `text`: text string from the document, with paragraphs separated by two newlines escape sequences. It is meant to be used directly as input for language modelling.
|
92 |
- `score`: positive float number representing the document quality, ranging from 0, which represents the worst quality, to 1, the best quality.
|
93 |
-
- `strategy`: text string describing the type of evaluation applied to obtain the document score.
|
94 |
- `languages`: dictionary containing the document languages, with a percentage indicating the character ratio for each one.
|
95 |
- `url`: text string with the URL of the document, if available.
|
96 |
|
|
|
78 |
|
79 |
### Data Instances
|
80 |
|
81 |
+
The dataset is provided in a JSONL format, where each row corresponds to a single document and contains a document identifier, the text, a quality score, the strategy used to evaluate the document quality, languages, and a URL of the document, if available.
|
82 |
|
83 |
```
|
84 |
+
{
|
85 |
+
"id": "macocu_ca_20230731_9_402472",
|
86 |
+
"text": "Jaume Casañas relleva Dolors Carreras a l’Alcaldia de l’Ajuntament de Cunit.
|
87 |
+
La substitució prevista al pacte de govern del 2019 s’ha materialitzat aquest
|
88 |
+
dissabte al matí. Aquest dissabte al matí, en un acte al Casal Municipal de
|
89 |
+
Cunit, s’ha celebrat l’acte de relleu de l’Alcaldia de l’Ajuntament de Cunit,
|
90 |
+
segons preveia el pacte de govern signat el juny del 2019 pels grups del PSC,
|
91 |
+
encapçalat per la fins ara alcaldessa, Dolors Carreras, i Impulsem Cunit, amb
|
92 |
+
el ja nou alcalde, Jaume Casañas, al capdavant.",
|
93 |
+
"score": 0.8105327621841463,
|
94 |
+
"strategy": "curate",
|
95 |
+
"languages": "{"ca": 1.0}",
|
96 |
+
"url": ""
|
97 |
+
}
|
98 |
```
|
99 |
|
100 |
### Data Fields
|
101 |
|
102 |
+
- `id`: text string containing the document identifier. Consists of the subdataset code, the part number and a document number.
|
103 |
- `text`: text string from the document, with paragraphs separated by two newlines escape sequences. It is meant to be used directly as input for language modelling.
|
104 |
- `score`: positive float number representing the document quality, ranging from 0, which represents the worst quality, to 1, the best quality.
|
105 |
+
- `strategy`: text string describing the type of evaluation applied to obtain the document score. "curate" uses the heuristic evaluation from [CURATE](https://github.com/langtech-bsc/corpus-cleaner-v2) and "perfect" means that manual review was done and the highest score (1) is applied.
|
106 |
- `languages`: dictionary containing the document languages, with a percentage indicating the character ratio for each one.
|
107 |
- `url`: text string with the URL of the document, if available.
|
108 |
|