metadata

license: cc
task_categories:
  - translation
language:
  - am
  - en
size_categories:
  - 10K<n<100K

AGE: Amharic, Ge’ez, and English Parallel Dataset

Overview

The AGE dataset is an open-source tripartite alignment of Amharic, Ge’ez, and English parallel sentences. This dataset aims to address the lack of high-quality resources for low-resource languages like Amharic and Ge’ez, facilitating the development of Natural Language Processing (NLP) models.

Data Collection

Our dataset was sourced from diverse materials, and we removed excessively disordered portions to ensure quality. The dataset development process is illustrated in the accompanying figure.

Below is an example row:

[
    {
        "amh": "ንጉሡ ዳዊትም ሸመገለ ዕድሜውም በዛ፤ ልብስም ደረቡለት፥ ነገር ግን አይሞቀውም ነበር።",
        "gez": "ወዳዊትሰ ንጉሥ ልህቀ ወኀለፈ መዋዕሊሁ ወይከድንዎ አልባሰ ወኢያመውቆ ።",
        "eng": "Now king David was old and stricken in years; and they covered him with clothes, but he gat no heat."
    },
    {
        "amh": "ባሪያዎቹም። ለጌታችን ለንጉሡ ድንግል ትፈለጋለች፤ በንጉሡም ፊት ቆማ ታገልግለው፥ በጌታችንም በንጉሡ ብብት ተኝታ ታሙቀው አሉት።",
        "gez": "ወይቤሉ ደቁ ለዳዊት ይኅሥሡ ለእግዚእነ ለንጉሥ ወለተ ድንግለ ወያምጽእዋ ኀበ ንጉሥ ወትሰክብ ምስሌሁ ወተሐቅፎ ወታስተማውቆ ወይመውቅ እግዚእነ ንጉሥ ።",
        "eng": "Wherefore his servants said unto him, Let there be sought for my lord the king a young virgin: and let her stand before the king, and let her cherish him, and let her lie in thy bosom, that my lord the king may get heat."
    },

]

መጽሐፈ ፡ ኩፋሌ (Book of Jubilees)

We have the Book of Jubilees only in Ge'ez and English

Kufale.xlsx - data and associated metadata in xlsx format.

Dataset Structure

The dataset is available in both CSV and JSON formats, with columns tagged for each language:

amh: Amharic
gez: Ge’ez
eng: English

Citation

If you use this dataset, please cite our paper:

@inproceedings{ademtew-birbo-2024-age,
    title = "{AGE}: {A}mharic, {G}e{'}ez and {E}nglish Parallel Dataset",
    author = "Ademtew, Henok  and
      Birbo, Mikiyas",
    editor = "Ojha, Atul Kr.  and
      Liu, Chao-hong  and
      Vylomova, Ekaterina  and
      Pirinen, Flammie  and
      Abbott, Jade  and
      Washington, Jonathan  and
      Oco, Nathaniel  and
      Malykh, Valentin  and
      Logacheva, Varvara  and
      Zhao, Xiaobing",
    booktitle = "Proceedings of the Seventh Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2024)",
    month = aug,
    year = "2024",
    address = "Bangkok, Thailand",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.loresmt-1.14",
    doi = "10.18653/v1/2024.loresmt-1.14",
    pages = "139--145",
    abstract = "African languages are not well-represented in Natural Language Processing (NLP). The main reason is a lack of resources for training models. Low-resource languages, such as Amharic and Ge{'}ez, cannot benefit from modern NLP methods because of the lack of high-quality datasets. This paper presents AGE, an open-source tripartite alignment of Amharic, Ge{'}ez, and English parallel dataset. Additionally, we introduced a novel, 1,000 Ge{'}ez-centered sentences sourced from areas such as news and novels. Furthermore, we developed a model from a multilingual pre-trained language model, which brings 12.29 and 30.66 for English-Ge{'}ez and Ge{'}ez to English, respectively, and 9.39 and 12.29 for Amharic-Ge{'}ez and Ge{'}ez-Amharic respectively.",
}