Hassan-IA/DAH

DAtaset Hassaniya # DAH (DAtaset Hassaniya) ## Project Overview **DAH (DAtaset Hassaniya)** is the first publicly available, open-source dataset for Hassaniya Arabic. Hassaniya is a dialect of Arabic spoken in Mauritania. Despite its widespread use, the language is not well-represented in digital resources. The primary goal of the DAH project is to bridge this gap. By creating a high-quality dataset, we aim to facilitate: 1. **Machine Translation:** The development of accurate translation tools between English and Hassaniya. This is crucial for improving technologies like online translators and other AI-driven language models. 2. **Linguistic Research:** Providing researchers with a structured dataset to study the grammar, vocabulary, and nuances of the Hassaniya dialect. 3. **Language Preservation:** Documenting Hassaniya in a digital format that is accessible to both the Hassaniya-speaking community and a global audience. *** ## Dataset Details The DAH dataset is structured as a parallel corpus, meaning that each entry consists of a sentence in English paired with its translation in Hassaniya. | Component | Description | | :--- | :--- | | `english` | The source sentence in English. | | `hassaniya-ar` | The Hassaniya translation written in the traditional Arabic script. | | `hassaniya-en` | The Hassaniya translation written in a Latin-based script, often referred to as "Arabizi." | * **Source:** The sentences are adapted from the Tatoeba Project (tatoeba.org) , a large, community-based collection of sentences and translations. * **Quality:** All entries have been reviewed by the founders of the Hassan-IA community to ensure accuracy. * **License:** The dataset is released under the CC-BY-4.0 license, which allows for open use and adaptation, provided that appropriate credit is given. *** ## Creators - Ahlam Abdelkader - Emani Babe - Oumoukelthoum Sidenna *** ## The Hassaniya Latin Script To maintain consistency, a standardized transliteration system …