Semi-automatic collation of versions of a text written in different languages

以不同语言写成的文本不是从源语到目标语的逐字和逐义翻译，而是文本的多语种版本，它们之间可能存在不同程度的差异，是许多学科研究的主题。可以找到以不同语言编写的文本的版本，例如，在以两种或更多语言编写其作品的作者的情况下，并且对其而言，语言之间的改变还可能涉及文本适应新的目标群体。该项目的目标是开发和实施一项电子研究技术，用于比较以不同语文编写的文本的非平行版本，并在示范应用环境中对其进行测试。该项目的成果将是一种软件，该软件能够识别在段落、句子、短语和单词层面上仅部分平行的文本之间的差异，并允许随后的交互式人工编辑。

Texts written in different languages which are not translations in the sense of a word-for-word and sense-for-sense transfer from a source language to a target language, but rather multilingual versions of a text, which can differ from one another to varying degrees, are the subject of research in numerous disciplines. Versions of a text written in different languages can be found, e.g., in the case of authors who write their works in two or more languages, and for whom the change between the languages also can involve an adaptation of the text to the new target group. The goal of this project is the development and implementation of an e-research technology for the comparison of non-parallel versions of a text written in different languages, and their testing in exemplary application contexts. The result of the project will be software that identifies the differences between only partially-parallel texts on paragraph, sentence, phrase and word level, and allows for subsequent interactive manual editing.