ParaCrawl

ParaCrawl是一个大规模平行语料库。它由欧盟通过“连接欧洲设施”计划共同资助，旨在为欧洲语言持续提供网络规模的平行语料。该数据库主要收录从多语言网站自动采集的句对齐文本，支持多种欧洲语言及部分其他语言对，并包含用于处理平行语料的Bitextor开源工具链。

ParaCrawl is a large-scale parallel corpus. It is jointly funded by the European Union through the Connecting Europe Facility program, with the objective of continuously providing web-scale parallel corpora for European languages. This corpus primarily consists of sentence-aligned texts automatically crawled from multilingual websites, supports multiple European languages and some additional language pairs, and includes the open-source Bitextor toolchain for processing parallel corpora.

搜集汇总

背景与挑战

背景概述

ParaCrawl是一个由欧盟资助的大规模平行语料库，旨在持续为欧洲语言提供网络规模的平行语料。它通过自动采集多语言网站的句对齐文本，支持多种欧洲语言及其他语言对，并包含用于处理平行语料的Bitextor开源工具链。