Encompassing the cleaned content of full Wikipedia articles, built from the Wikipedia dump. Each example is a full article stripped of markdown and unwanted sections like references. The row count, column count, and specific cleaning methods are not detailed.
Use Cases
- Train a language model on full Wikipedia article content for text generation tasks
- Analyze the structure and topics of cleaned Wikipedia articles for content categorization
- Use the stripped article text as a corpus for information retrieval or semantic search systems
Strengths
- Derived from the authoritative Wikipedia dump source
- Includes cleaning to remove markdown and unwanted sections like references
- Contains the full content of Wikipedia articles
Limitations
- The specific cleaning process and criteria for removed sections are not detailed
- The dataset's size, row count, and update frequency relative to Wikipedia are unknown
- Potential biases or completeness of the Wikipedia source are not addressed
Provenance
- Source
- Wikipedia dump (https://dumps.wikimedia.org/)
- Collection Method
- Built from the dump with cleaning to strip markdown and unwanted sections
- Time Range
- null
- Freshness
- null
- Geography
- null