Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -39,4 +39,58 @@ task_categories:
|
|
| 39 |
task_ids:
|
| 40 |
- multiple-choice-qa
|
| 41 |
---
|
| 42 |
-
MFAQ
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
task_ids:
|
| 40 |
- multiple-choice-qa
|
| 41 |
---
|
| 42 |
+
# MFAQ
|
| 43 |
+
MFAQ is a multilingual corpus of *Frequently Asked Questions* obtained by scraping the [Common Crawl](https://commoncrawl.org/).
|
| 44 |
+
|
| 45 |
+
## Languages
|
| 46 |
+
We collected data in 21 different languages. To download a language specific subset you need to specify the language key as configuration. See below for an example.
|
| 47 |
+
```
|
| 48 |
+
load_dataset("clips/mfaq", "en")
|
| 49 |
+
```
|
| 50 |
+
|
| 51 |
+
| Language | Key | Pairs | Pages |
|
| 52 |
+
|------------|-----|-----------|-----------|
|
| 53 |
+
| All | all | 6,346,693 | 1,035,649 |
|
| 54 |
+
| English | en | 3,719,484 | 608,796 |
|
| 55 |
+
| German | de | 829,098 | 111,618 |
|
| 56 |
+
| Spanish | es | 482,818 | 75,489 |
|
| 57 |
+
| French | fr | 351,458 | 56,317 |
|
| 58 |
+
| Italian | it | 155,296 | 24,562 |
|
| 59 |
+
| Dutch | nl | 150,819 | 32,574 |
|
| 60 |
+
| Portuguese | pt | 138,778 | 26,169 |
|
| 61 |
+
| Turkish | tr | 102,373 | 19,002 |
|
| 62 |
+
| Russian | ru | 91,771 | 22,643 |
|
| 63 |
+
| Polish | pl | 65,182 | 10,695 |
|
| 64 |
+
| Indonesian | id | 45,839 | 7,910 |
|
| 65 |
+
| Norwegian | no | 37,711 | 5,143 |
|
| 66 |
+
| Swedish | sv | 37,003 | 5,270 |
|
| 67 |
+
| Danish | da | 32,655 | 5,279 |
|
| 68 |
+
| Vietnamese | vi | 27,157 | 5,261 |
|
| 69 |
+
| Finnish | fi | 20,485 | 2,795 |
|
| 70 |
+
| Romanian | ro | 17,066 | 3,554 |
|
| 71 |
+
| Czech | cs | 16,675 | 2,568 |
|
| 72 |
+
| Hebrew | he | 11,212 | 1,921 |
|
| 73 |
+
| Hungarian | hu | 8,598 | 1,264 |
|
| 74 |
+
| Croatian | hr | 5,215 | 819 |
|
| 75 |
+
|
| 76 |
+
## Data Fields
|
| 77 |
+
The dataset is organized by page. Each page contains a list of questions and answers. You can also retrieve a flattened representation (see the next section).
|
| 78 |
+
- **id**
|
| 79 |
+
- **language**
|
| 80 |
+
- **num_pairs**: the number of FAQs on the page
|
| 81 |
+
- **domain**: source web domain of the FAQs
|
| 82 |
+
- **qa_pairs**: a list of questions and answers
|
| 83 |
+
- **question**
|
| 84 |
+
- **answer**
|
| 85 |
+
- **language**
|
| 86 |
+
|
| 87 |
+
## Flat Representation
|
| 88 |
+
You can access the flat version of any language by appending `_flat` to the configuration (e.g. `en_flat`). The data will be returned pair-by-pair instead of page-by-page.
|
| 89 |
+
|
| 90 |
+
## Source Data
|
| 91 |
+
|
| 92 |
+
This section was adapted from the source data description of [OSCAR](https://huggingface.co/datasets/oscar#source-data)
|
| 93 |
+
|
| 94 |
+
Common Crawl is a non-profit foundation which produces and maintains an open repository of web crawled data that is both accessible and analysable. Common Crawl's complete web archive consists of petabytes of data collected over 8 years of web crawling. The repository contains raw web page HTML data (WARC files), metdata extracts (WAT files) and plain text extracts (WET files). The organisation's crawlers has always respected nofollow and robots.txt policies.
|
| 95 |
+
|
| 96 |
+
To construct MFAQ, the WARC files of Common Crawl were used. We looked for `FAQPage` markup in the HTML and subsequently parsed the `FAQItem` from the page.
|