license: other
license_name: bright-data-master-service-agreement
license_link: https://brightdata.com/license
language:
- en
task_categories:
- text-classification
- text-generation
- text2text-generation
- summarization
- question-answering
tags:
- wikipedia
- text
- NLP
- ML
- AI
- Knowledge Extraction
- Natural Language Processing
- Information Retrieval
- LLM
Dataset Card for "BrightData/Wikipedia-Articles"
If you are using this dataset, we would love your feedback: Link to form.
Dataset Summary
Explore a collection of millions of Wikipedia articles with the Wikipedia dataset, comprising over 1.23M structured records and 10 data fields updated and refreshed regularly.
Each entry includes all major data points such as including timestamps, URLs, article titles, raw and cataloged text, images, "see also" references, external links, and a structured table of contents.
For a complete list of data points, please refer to the full "Data Dictionary" provided below.
To explore additional free and premium datasets, visit our website brightdata.com.
Data Dictionary
| Column name | Description | Data type |
|---|---|---|
| url | URL of the article | Url |
| title | Title of the article | Text |
| table_of_contents | Table of Contents in the article | Array |
| raw_text | Raw article text | Text |
| cataloged_text | Cataloged text of the article by titles | Array |
| > title | Title of a cataloged section | Text |
| > sub_title | Subtitle within a cataloged section | Text |
| > text | Text content within a cataloged section | Text |
| > links_in_text | Links within the text content | Array |
| >> link_name | Name or description of the link | Text |
| >> url | URL of the link | Url |
| images | Links to the URLs of images in the article | Array |
| > image_text | Text description under an image | Text |
| > image_url | URL of the image | Url |
| see_also | Other recommended articles | Array |
| > title | Recommended article title | Text |
| > url | URL of the recommended article | Url |
| references | References in the article | Array |
| > reference | Reference in the article | Text |
| >> urls | URLs referenced within the article | Array |
| >>> url_text | Text description of the referenced URL | Text |
| >>> url | URL of the referenced article or source | Url |
| external_links | External links referenced in the article | Array |
| > external_links_name | Name or description of the external link | Text |
| > link | External link URL | Text |
Dataset Creation
Data Collection and Processing
The data collection process involved extracting information directly from Goodreads, ensuring comprehensive coverage of the required attributes. Once collected, the data underwent several stages of processing:
- Parsing: Extracted raw data was parsed to convert it into a structured format.
- Cleaning: The cleaning process involved removing any irrelevant or erroneous entries to enhance data quality.
Validation
To ensure data integrity, a validation process was implemented. Each entry is checked across various attributes, including:
- Uniqueness: Each record was checked to ensure it was unique, eliminating any duplicates.
- Completeness: The dataset was examined to confirm that all necessary fields were populated or filled, with missing data addressed appropriately.
- Consistency: Cross-validation checks were conducted to ensure consistency across various attributes, including comparison with historical records.
- Data Types Verification: Ensured that all data types were correctly assigned and consistent with expected formats.
- Fill Rates and Duplicate Checks: Conducted comprehensive checks to verify fill rates, ensuring no significant gaps in data, and rigorously screened for duplicates. This ensures that the dataset meets the high standards of quality necessary for analysis, research and modeling.