Amharic health question answering dataset

The **AmHQA** dataset is an Amharic health question answering corpus curated to support research in low-resource language natural language processing and medical question answering. The dataset consists of **1600 question–answer pairs for training** and **400 pairs for testing**, all provided in **CSV format**. The content is written entirely in **Amharic** and is intended to facilitate the development and evaluation of extractive and neural question answering systems in the health domain. AmHQA is released under the **Creative Commons Attribution 4.0 International (CC BY 4.0) licence**, allowing unrestricted use, distribution, and adaptation with appropriate attribution. Researchers using this dataset are requested to cite it as: *Bogale, B., et al. (2026). AmHQA: An Amharic Health Question Answering Dataset. Mendeley Data*. For further information or inquiries, please contact **[[e-mail]](mailto:[e-mail])** (mobile: **0938282528**).