Med-EASi: 1,979 Expert-Simple Medical Text Pairs for Supervised Simplification

Med-EASi is a crowdsourced dataset containing 1,979 expert-simple text pairs for supervised simplification of short medical texts. It was created by Chandrayee Basu and associated with a paper published on arXiv in 2023. The dataset was last updated on the Hugging Face platform on March 8, 2023.

Use Cases

• Train supervised text simplification models based on the expert-simple text pairs.
• Benchmark abstractive summarization models for medical content.
• Develop models for improving health literacy by simplifying complex medical language.
• Study the linguistic transformations between expert and layperson medical terminology.

Strengths

• Contains 1,979 finely annotated expert-simple text pairs.
• Specifically designed for the medical domain, a niche for text simplification.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last updated 2023-03-08 18:24:31; freshness should be verified.

Provenance

Source

Hugging Face dataset repository by author 'cbasu'.

Collection Method

Crowdsourced and finely annotated.

Freshness

Last updated 2023-03-08 18:24:31.