docs: Update readme

README.md

@@ -1,100 +1,69 @@
 ---
-pretty_name: ScandiQA
+pretty_name: ScandiWiki
 language:
 - da
 - sv
 - no
+- nb
+- nn
+- is
+- fo
 license:
 - cc-by-sa-4.0
 multilinguality:
 - multilingual
 size_categories:
-- 1K<n<10K
+- 1M<n<10M
 source_datasets:
-- mkqa
-- natural_questions
+- wikipedia
 task_categories:
-- question-answering
+- fill-mask
+- text-generation
+- feature-extraction
 task_ids:
-- extractive-qa
+- language-modeling
 ---
 
-# Dataset Card for ScandiQA
+# Dataset Card for ScandiWiki
 
 ## Dataset Description
 
-- **Repository:** <https://github.com/alexandrainst/scandi-qa>
 - **Point of Contact:** [Dan Saattrup Nielsen](mailto:dan.nielsen@alexandra.dk)
-- **Size of downloaded dataset files:** 69 MB
-- **Size of the generated dataset:** 67 MB
-- **Total amount of disk used:** 136 MB
+- **Size of downloaded dataset files:** 4200 MB
+- **Size of the generated dataset:** 4200 MB
+- **Total amount of disk used:** 8400 MB
 
 ### Dataset Summary
 
-ScandiQA is a dataset of questions and answers in the Danish, Norwegian, and Swedish
-languages. All samples come from the Natural Questions (NQ) dataset, which is a large
-question answering dataset from Google searches. The Scandinavian questions and answers
-come from the MKQA dataset, where 10,000 NQ samples were manually translated into,
-among others, Danish, Norwegian, and Swedish. However, this did not include a
-translated context, hindering the training of extractive question answering models.
-
-We merged the NQ dataset with the MKQA dataset, and extracted contexts as either "long
-answers" from the NQ dataset, being the paragraph in which the answer was found, or
-otherwise we extract the context by locating the paragraphs which have the largest
-cosine similarity to the question, and which contains the desired answer.
-
-Further, many answers in the MKQA dataset were "language normalised": for instance, all
-date answers were converted to the format "YYYY-MM-DD", meaning that in most cases
-these answers are not appearing in any paragraphs. We solve this by extending the MKQA
-answers with plausible "answer candidates", being slight perturbations or translations
-of the answer.
-
-With the contexts extracted, we translated these to Danish, Swedish and Norwegian using
-the [DeepL translation service](https://www.deepl.com/pro-api?cta=header-pro-api) for
-Danish and Swedish, and the [Google Translation
-service](https://cloud.google.com/translate/docs/reference/rest/) for Norwegian. After
-translation we ensured that the Scandinavian answers do indeed occur in the translated
-contexts.
-
-As we are filtering the MKQA samples at both the "merging stage" and the "translation
-stage", we are not able to fully convert the 10,000 samples to the Scandinavian
-languages, and instead get roughly 8,000 samples per language. These have further been
-split into a training, validation and test split, with the latter two containing
-roughly 750 samples. The splits have been created in such a way that the proportion of
-samples without an answer is roughly the same in each split.
-
+ScandiWiki is a parsed and deduplicated Wikipedia dump in Danish, Norwegian Bokmål,
+Norwegian Nynorsk, Swedish, Icelandic and Faroese.
 
 ### Supported Tasks and Leaderboards
 
-Training machine learning models for extractive question answering is the intended task
-for this dataset. No leaderboard is active at this point.
+This dataset is intended for general language modelling.
 
 
 ### Languages
 
-The dataset is available in Danish (`da`), Swedish (`sv`) and Norwegian (`no`).
+The dataset is available in Danish (`da`), Swedish (`sv`), Norwegian Bokmål (`nb`),
+Norwegian Nynorsk (`nn`), Icelandic (`is`) and Faroese (`fo`).
 
 
 ## Dataset Structure
 
 ### Data Instances
 
-- **Size of downloaded dataset files:** 69 MB
-- **Size of the generated dataset:** 67 MB
-- **Total amount of disk used:** 136 MB
+- **Size of downloaded dataset files:** 4200 MB
+- **Size of the generated dataset:** 4200 MB
+- **Total amount of disk used:** 8400 MB
 
-An example from the `train` split of the `da` subset looks as follows.
+An example from the `train` split of the `fo` subset looks as follows.
 ```
 {
-    'example_id': 123,
-    'question': 'Er dette en test?',
-    'answer': 'Dette er en test',
-    'answer_start': 0,
-    'context': 'Dette er en testkontekst.',
-    'answer_en': 'This is a test',
-    'answer_start_en': 0,
-    'context_en': "This is a test",
-    'title_en': 'Train test'
+    'id': '3380',
+    'url': 'https://fo.wikipedia.org/wiki/Enk%C3%B6pings%20kommuna',
+    'title': 'Enköpings kommuna',
+    'text': 'Enköpings kommuna (svenskt: Enköpings kommun), er ein kommuna í Uppsala län í Svøríki. Enköpings kommuna hevur umleið 40.656 íbúgvar (2013).\n\nKeldur \n\nKommunur í Svøríki'
 }
 ```
 
@@ -102,38 +71,34 @@ An example from the `train` split of the `da` subset looks as follows.
 
 The data fields are the same among all splits.
 
-- `example_id`: an `int64` feature.
-- `question`: a `string` feature.
-- `answer`: a `string` feature.
-- `answer_start`: an `int64` feature.
-- `context`: a `string` feature.
-- `answer_en`:  a `string` feature.
-- `answer_start_en`: an `int64` feature.
-- `context_en`: a `string` feature.
-- `title_en`: a `string` feature.
+- `id`: a `string` feature.
+- `url`: a `string` feature.
+- `title`: a `string` feature.
+- `text`: a `string` feature.
 
-### Data Splits
+### Data Subsets
 
-|   name   | train | validation | test |
-|----------|------:|-----------:|-----:|
-| da       | 6311  | 749        | 750  |
-| sv       | 6299  | 750        | 749  |
-| no       | 6314  | 749        | 750  |
+|   name   | samples   |
+|----------|----------:|
+| da       | 287,216   |
+| sv       | 2,469,978 |
+| nb       | 596,593   |
+| nn       | 162,776   |
+| is       | 55,418    |
+| fo       | 12,582    |
 
 
 ## Dataset Creation
 
 ### Curation Rationale
 
-The Scandinavian languages does not have any gold standard question answering dataset.
-This is not quite gold standard, but the fact both the questions and answers are all
-manually translated, it is a solid silver standard dataset.
+It takes quite a long time to parse the Wikipedia dump as well as to deduplicate it, so
+this dataset is primarily for convenience.
 
 ### Source Data
 
-The original data was collected from the [MKQA](https://github.com/apple/ml-mkqa/) and
-[Natural Questions](https://ai.google.com/research/NaturalQuestions) datasets from
-Apple and Google, respectively.
+The original data is from the [wikipedia
+dataset](https://huggingface.co/datasets/wikipedia).
 
 
 ## Additional Information
@@ -146,4 +111,5 @@ Institute](https://alexandra.dk/) curated this dataset.
 ### Licensing Information
 
 The dataset is licensed under the [CC BY-SA 4.0
-license](https://creativecommons.org/licenses/by-sa/4.0/).
+license](https://creativecommons.org/licenses/by-sa/4.0/), in accordance with the same
+license of the [wikipedia dataset](https://huggingface.co/datasets/wikipedia).