Domain-specific multilingual translation for producers of information.

EAMT(2016)

引用 0|浏览10
暂无评分
摘要
Producers of multilingual information (webpages, manuals, etc) often need not solve the general problem of “translating anything”, but can stay content with a more limited system, which however must be accurate. One way to do this is to define a semantic interlingua, which exactly fits the content to be translated. It has turned out that algebraic datatypes, and the way they are used in compilers to encode abstract syntax, are useful for this purpose. The grammar formalism Grammatical Framework (GF) was originally designed to support this kind of systems. While much of the current research in GF is focused on scaling it up to general-purpose translation, the creation of domain-specific systems has become a routine task, which is commercially exploited by the start-up company Digital Grammars AB. In the conference, we plan to show a demo of the tools for building such systems, as well as some actual systems built for customers.
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要