NaijaRC

NaijaRC是一个专为尼日利亚语言设计的多项选择阅读理解数据集，涵盖了Hausa、Igbo和Yorùbá三种语言。该数据集由Masakhane NLP创建，基于高中阅读理解考试题目，旨在提升机器在理解长段落和回答问题方面的性能。数据集内容包括阅读理解段落、问题及答案，均由母语者和语言学家仔细验证和清理。NaijaRC的应用领域主要集中在提升非洲语言的机器阅读理解能力，解决资源匮乏语言的AI模型开发挑战。

NaijaRC is a multiple-choice reading comprehension dataset tailored for Nigerian languages, encompassing Hausa, Igbo, and Yorùbá. It was developed by Masakhane NLP and sourced from high school reading comprehension examination questions, with the primary objective of advancing machine performance in long passage comprehension and question answering. The dataset comprises reading comprehension passages, questions, and corresponding answers, all of which have been rigorously verified and curated by native speakers and linguists. The core application scope of NaijaRC lies in improving machine reading comprehension capabilities for African languages, and addressing the development hurdles of AI models for low-resource languages.