Wikipedia Comments

CHANGELOG.md

@@ -5,6 +5,12 @@ All notable changes to this project will be documented in this file.
 
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
+## [v1.2.6] - 2025-07-21
+
+### Added
+
+- Added the `wiki-comments` dataset.
+
 ## [v1.2.5] - 2025-07-08
 
 ### Added

CONTRIBUTING.md

@@ -42,6 +42,12 @@ This repo comes with a few dependencies you need to install to make this run. It
 make install
 ```
 
+Now you can activate the environment with:
+
+```
+source .venv/bin/activate
+```
+
 ## Running dataset tests
 
 This dataset is special as it comes with a test suite, e.g. testing in the ids are unique and that the format is consistent. You can run the suite using
@@ -96,4 +102,4 @@ We will for instance examine the quality of the synthetic subset and whether the
 
 ### Do you accept non-Danish data
 
-Generally this repository is intended for Danish text, however quite broadly defined. For instance, we do accept data containing [code-switching](https://www.google.com/search?client=safari&rls=en&q=code+switching&ie=UTF-8&oe=UTF-8) and historical Danish text.
+Generally this repository is intended for Danish text, however quite broadly defined. For instance, we do accept data containing [code-switching](https://www.google.com/search?client=safari&rls=en&q=code+switching&ie=UTF-8&oe=UTF-8) and historical Danish text.

README.md

@@ -129,6 +129,10 @@ configs:
   data_files:
   - split: train
     path: data/wiki/*.parquet
+- config_name: wiki-comments
+  data_files:
+  - split: train
+    path: data/wiki-comments/*.parquet
 - config_name: nordjyllandnews
   data_files:
   - split: train
@@ -182,7 +186,7 @@ https://github.com/huggingface/datasets/blob/main/templates/README_guide.md
 <!-- START README TABLE -->
 |              |                                                                                                                                                             |
 | ------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| **Version** | 1.2.5 ([Changelog](/CHANGELOG.md)) |
+| **Version** | 1.2.6 ([Changelog](/CHANGELOG.md)) |
 | **Language** | dan, dansk, Danish                                                                                                                                          |
 | **License**  | Openly Licensed, See the respective dataset                                                                                                                 |
 | **Models**   | For model trained used this data see [danish-foundation-models](https://huggingface.co/danish-foundation-models)                                            |

data/wiki-comments/create.py

@@ -0,0 +1,119 @@
+# /// script
+# requires-python = "==3.12"
+# dependencies = [
+#     "datasets==3.2.0",
+#     "dynaword",
+#     "fasttext",
+#     "huggingface_hub"
+# ]
+# [tool.uv.sources]
+# dynaword = { git = "https://huggingface.co/datasets/danish-foundation-models/danish-dynaword", rev = "6b3822fd6965dda59ae361da99c19b5c56c1263f" }
+# ///
+"""
+This script downloads and cleans Wikipedia, it only keeps the user comments. 
+"""
+import html
+import json
+import os
+import requests
+import sys
+
+from datasets import IterableDataset
+from datasets import Dataset
+import fasttext
+from huggingface_hub import hf_hub_download
+
+from dynaword.process_dataset import (
+    add_token_count,
+    ensure_column_order,
+    remove_duplicate_text,
+    remove_empty_texts,
+)
+
+
+def run_cmd(cmd):
+    print(cmd)
+    os.system(cmd)
+
+def download_data(lang, date):
+    filename = lang + 'wiki-' + date + '-pages-articles.xml.bz2'
+    if not os.path.isfile(filename):
+        url = 'https://dumps.wikimedia.org/' + lang + 'wiki/' + date + '/' + filename
+        response = requests.get(url)
+
+        with open(filename, "wb") as file:
+            file.write(response.content)
+
+    return filename
+
+def install_wikiextractor():
+    if not os.path.isdir('wikiextractor'):
+        cmd = 'git clone https://github.com/robvanderg/wikiextractor.git'
+        run_cmd(cmd)
+
+def run_wikiextractor(in_path, out_path):
+    # clean the data
+    if not os.path.isdir('wikiextractor/' + out_path):
+        cmd = 'cd wikiextractor && python3 -m wikiextractor.WikiExtractor ../' + in_path + ' -o ' + out_path + ' --get_misc --json && cd ../ '
+        run_cmd(cmd)
+
+def read_and_clean(path):
+    # Load fasttext model
+    model_path = hf_hub_download(repo_id="facebook/fasttext-language-identification", filename="model.bin")
+    fasttext_model = fasttext.load_model(model_path)
+
+    comment_id = 0
+    all_rows = []
+    for (root,dirs,files) in os.walk(path, topdown=True):
+        for file in files:
+            path = os.path.join(root, file)
+            for line in open(path):
+                linedata = json.loads(line)
+                title = linedata['title']
+                category = title.split(':')[0]
+                if category == 'Wikipedia':
+                    if title.startswith('Wikipedia:Dagens '):
+                        continue
+                    id = 'wikicomment_' + str(comment_id)
+                    comment_id += 1
+                else: # There is more data, but we just want to comments for now
+                    continue
+                source = 'wiki_misc'
+                # TODO add linedata['url'] somewhere?
+                text = html.unescape(linedata['text'])
+                lines = line.split('\n')
+                filtered_text = ''
+                for line in text.split('\n'):
+                    if '{{}}' in text: # unresolved templates
+                        continue
+
+                    lang_pred = fasttext_model.predict(line)
+                    if lang_pred[0][0] == '__label__dan_Latn' and lang_pred[1][0] > .5:    
+                        filtered_text += line + '\n'
+                added = '2025-07-21'
+                created = '2002-02-01, 2025-07-20'
+                row = {"id": id, "text": filtered_text, "source": source, "added": added, "created": created}
+                all_rows.append(row)
+    return all_rows
+     
+
+if __name__ == '__main__':
+    date = '20250720' # obtained from https://dumps.wikimedia.org/dawiki/
+    lang = 'da'
+    bz2_path = download_data(lang, date)
+
+    install_wikiextractor()
+    data_folder = lang + 'wiki-misc-' + date
+    run_wikiextractor(bz2_path, data_folder)
+
+    full_data = read_and_clean('wikiextractor/' + data_folder)
+
+    ds = Dataset.from_list(full_data)
+
+    ds = remove_empty_texts(ds)
+    ds = remove_duplicate_text(ds)
+    ds = add_token_count(ds)
+    ds = ensure_column_order(ds)
+
+    ds.to_parquet('data/wiki-comments/wiki-comments.parquet')
+

data/wiki-comments/descriptive_stats.json

@@ -0,0 +1,6 @@
+{
+  "number_of_samples": 12462,
+  "average_document_length": 2101.6405071417107,
+  "number_of_tokens": 8805990,
+  "revision": "6b3822fd6965dda59ae361da99c19b5c56c1263f"
+}

data/wiki-comments/wiki-comments.md

@@ -0,0 +1,97 @@
+---
+pretty_name: Wikipedia Comments
+language:
+- da
+license: cc0-1.0
+license_name: CC-0
+size_categories:
+- 100k-1M
+task_categories:
+- text-generation
+- fill-mask
+task_ids:
+- language-modeling
+domains:
+- Encyclopedic
+---
+
+# Dataset Card for Wikipedia Comments
+
+<!-- START-SHORT DESCRIPTION -->
+Text from the comments sections of the Danish Wikipedia.
+<!-- END-SHORT DESCRIPTION -->
+
+
+You can read more about the wikipedia on their [about](https://da.wikipedia.org/wiki/Hj%C3%A6lp:Om) page.
+
+## Dataset Description
+
+
+<!-- START-DESC-STATS -->
+- **Language**: dan, dansk, Danish
+- **Domains**: Encyclopedic
+- **Number of samples**: 12.46K
+- **Number of tokens (Llama 3)**: 8.81M
+- **Average document length (characters)**: 2101.64
+<!-- END-DESC-STATS -->
+
+
+
+## Dataset Structure
+An example from the dataset looks as follows.
+
+
+<!-- START-SAMPLE -->
+```py
+{
+  "id": "wikicomment_0",
+  "text": " A B C D E F G H I J K L M N O P Q R S T U V W X Y Z Æ Ø Å\nDette er en noget-nær-alfabetisk liste ov[...]",
+  "source": "wiki_misc",
+  "added": "2025-07-21",
+  "created": "2002-02-01, 2025-07-20",
+  "token_count": 161
+}
+```
+
+### Data Fields
+
+An entry in the dataset consists of the following fields:
+
+- `id` (`str`): An unique identifier for each document.
+- `text`(`str`): The content of the document.
+- `source` (`str`): The source of the document (see [Source Data](#source-data)).
+- `added` (`str`): An date for when the document was added to this collection.
+- `created` (`str`): An date range for when the document was originally created.
+- `token_count` (`int`): The number of tokens in the sample computed using the Llama 8B tokenizer
+<!-- END-SAMPLE -->
+
+### Dataset Statistics
+
+<!-- START-DATASET PLOTS -->
+<p align="center">
+<img src="./images/dist_document_length.png" width="600" style="margin-right: 10px;" />
+</p>
+<!-- END-DATASET PLOTS -->
+
+
+
+## Additional Information
+
+This dataset is collected using an adapted version of the [WikiExtractor](https://github.com/attardi/wikiextractor). Rob van der Goot created a fork that allows for extracting additional text from Wiki's. The fork can be found here: [WikiExtractor](https://github.com/robvanderg/wikiextractor.git). 
+
+After inspection of the different outputs, there are multiple categories of files, which can most easily be distinguished through the title field. Below, I list the different categories, their size (number of pages), and what they seem to contain after a manual inspection. 
+
+```
+  71472 Kategori: category overview pages
+  19992 Wikipedia: Comments, but also daily articles
+   2379 Portal: Also monthly articles, and some lists/calendars 
+   1360 MediaWiki: About files, contains almost no natural language
+    726 Modul: technical stuff, contains almost no (Danish) text
+    171 Hjælp: help pages; info and comments
+```
+
+In the current version of the dataset, we used the titles starting with `Wikipedia:` , and remove the daily articles by leaving out titles starting with "Wikipedia:Dagens".
+
+In this data we include comments where people discuss things like: content of pages, writing style, which pages/information to include/exclude, etc. It also includes pages written for people that contribute to Wikipedia.  
+
+### Citation Information

data/wiki-comments/wiki-comments.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64c8b39dcd96ef6ac2bcfb5403cf9cd74f2ee17b954de3478f543922ee8dea5e
+size 10828848

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "dynaword"
-version = "1.2.5"
+version = "1.2.6"
 description = "project code for the danish dynaword project"
 readme = "README.md"
 requires-python = ">=3.12,<3.13" # 3.13 have issues with spacy and pytorch

test_results.log

@@ -1,25 +1,24 @@
 ============================= test session starts ==============================
-platform darwin -- Python 3.12.0, pytest-8.3.4, pluggy-1.5.0
-rootdir: /Users/kristianjensen/Documents/danish-dynaword
+platform linux -- Python 3.12.3, pytest-8.3.4, pluggy-1.5.0
+rootdir: /home/rob/Projects/danish-dynaword
 configfile: pyproject.toml
-plugins: anyio-4.9.0
-collected 328 items
+collected 337 items
 
 src/tests/test_dataset_schema.py ....................................... [ 11%]
-.................................                                        [ 21%]
-src/tests/test_datasheets.py ........................................... [ 35%]
-........................................................................ [ 57%]
-.................................................................        [ 76%]
+...................................                                      [ 21%]
+src/tests/test_datasheets.py ........................................... [ 34%]
+........................................................................ [ 56%]
+......................................................................   [ 76%]
 src/tests/test_load.py ..                                                [ 77%]
 src/tests/test_quality/test_duplicates.py .............................. [ 86%]
-......s                                                                  [ 88%]
+.......s                                                                 [ 88%]
 src/tests/test_quality/test_short_texts.py ............................. [ 97%]
-.......                                                                  [ 99%]
+........                                                                 [ 99%]
 src/tests/test_unique_ids.py .                                           [100%]
 
 =============================== warnings summary ===============================
-src/tests/test_quality/test_short_texts.py: 36 warnings
-  /Users/kristianjensen/Documents/danish-dynaword/.venv/lib/python3.12/site-packages/datasets/utils/_dill.py:385: DeprecationWarning: co_lnotab is deprecated, use co_lines instead.
+src/tests/test_quality/test_short_texts.py: 37 warnings
+  /home/rob/Projects/danish-dynaword/.venv/lib/python3.12/site-packages/datasets/utils/_dill.py:385: DeprecationWarning: co_lnotab is deprecated, use co_lines instead.
 
 -- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
-================= 327 passed, 1 skipped, 36 warnings in 52.74s =================
+================= 336 passed, 1 skipped, 37 warnings in 44.95s =================