Document Summarization Dataset from AI Hub

Encompassing text data for document summarization tasks, sourced from AI Hub. The dataset size is categorized as between 10,000 and 100,000 entries, and it was last updated in December 2021.

Use Cases

• Fine-tune a text summarization model on the provided document-summary pairs.
• Analyze the relationship between source document length and generated summary length.
• Benchmark extractive or abstractive summarization algorithms using the provided text data.

Strengths

• Dataset size is categorized as between 10,000 and 100,000 entries, providing a substantial corpus for training.
• Data is specifically curated for the document summarization task, offering a focused application.

Limitations

• The dataset was last updated in December 2021, which may limit its relevance for contemporary language model training.
• Specific column structure, sample data, and detailed metadata are unavailable, complicating initial assessment.

Provenance

Source

AI Hub

Freshness

Last updated 2021-12-14.

Geography

Region is indicated as 'us'.