FrancophonIA/DGT-Acquis

DGT-Acquis是一个由多个多语言平行语料库组成的家族，这些语料库是从欧盟官方期刊（OJ）中提取的，时间跨度为2004年中期至2011年底，涵盖多达23种语言。原始数据经过多个步骤的处理，每个步骤都会将前一步的结果细化到更精细的粒度。处理结果以自包含的多语言数据集格式（muset）文件打包。这些muset文件虽然独立，但相互之间存在链接，可以找到任何给定文本段落的源文档。数据用户可以根据自己的需要选择最合适处理级别的数据。原始数据包括XML和tiff文件，可用于其他类型的应用（如光学字符识别等）。文件级别格式适用于需要访问完整文本的用户，段落级别格式适用于不需要访问完整文本，但主要对更小段落的翻译感兴趣的用户。

The DGT-Acquis is a family of several multilingual parallel corpora extracted from the Official Journal of the European Union (OJ) between mid-2004 and the end of 2011, covering up to 23 languages. The original data has undergone several processing steps, each refining the result from the previous step to a finer granularity. The outcome of each step is packaged as a self-contained Multilingual Dataset Format (muset) file. These muset files are independent yet interconnected, allowing users to trace back to the source document of any given text segment. Users can select data at the processing level that best suits their needs. The original data includes both XML and tiff files, which can be used for other types of applications such as optical character recognition. The file level formats are relevant for users who need access to the full texts, while the paragraph level format is for those interested in smaller segments and their translations.