ITU Faroese Pairs Dataset
收藏arXiv2022-06-17 更新2024-06-21 收录
下载链接：
https://huggingface.co/datasets/strombergnlp/itu_faroese_danish
下载链接
链接失效反馈官方服务：
资源简介：
ITU Faroese Pairs Dataset是由哥本哈根信息技术大学创建的一个包含4000对法罗语和丹麦语句子的数据集。该数据集主要用于训练机器翻译系统，以提高法罗语的翻译质量。数据集的内容包括从法罗语报纸、欧洲议会记录和Tatoeba多语种数据集中提取的句子。创建过程中，数据集采用了多种来源的文本，并由母语为法罗语的专家进行翻译。该数据集的应用领域主要集中在改善法罗语与丹麦语之间的机器翻译，解决法罗语资源稀缺的问题。
ITU Faroese Pairs Dataset is a dataset containing 4000 pairs of Faroese and Danish sentences, developed by the IT University of Copenhagen. This dataset is primarily used for training machine translation systems to improve the translation quality for Faroese. The dataset's content consists of sentences extracted from Faroese newspapers, European Parliament proceedings, and the Tatoeba multilingual dataset. During its creation, texts from multiple sources were utilized, and the translations were completed by experts who are native Faroese speakers. Its main application focuses on enhancing machine translation between Faroese and Danish, addressing the problem of scarce Faroese language resources.
提供机构：
哥本哈根信息技术大学
创建时间：
2022-06-17
