NFCorpus

NFCorpus 是用于医学信息检索的全文英文检索数据集。它总共包含 3,244 个自然语言查询（用非技术英语编写，从 NutritionFacts.org 网站收集），其中 169,756 个自动提取的相关性判断对 9,964 个医学文档（用复杂的术语繁重的语言编写），主要来自 PubMed。

NFCorpus is a full-text English retrieval dataset dedicated to medical information retrieval. It contains a total of 3,244 natural language queries, which are written in non-technical English and collected from the NutritionFacts.org website. Additionally, it includes 169,756 automatically extracted relevance judgment pairs corresponding to 9,964 medical documents. These documents are written in complex, terminology-heavy language and are primarily sourced from PubMed.

搜集汇总

背景与挑战

背景概述

NFCorpus 是一个医学信息检索领域的英文全文检索数据集，包含 3,244 个自然语言查询和 169,756 个相关性判断，这些判断基于 9,964 篇来自 PubMed 等来源的医学术语密集文档。