Figshare · 2018 · 79 citations · 1 references
ComputingComputer ScienceLanguage DocumentationNatural Language ProcessingTranslation TheoryMachine TranslationMultilingual Natural Language ProcessingTranslation TechnologyDomain Knowledge AcquisitionMachine-assisted TranslationMultilingual Document ProcessingMultilingual Machine TranslationTranslation Automation
Knowledge-based interlingual machine translation systems produce semantically accurate translations, but typically require massive knowledge acquisition. This paper describes KANT, a system that reduces this requirement to produce practical, scalable, and accurate KBMT applications. First, the set of requirements is discussed, then the full KANT architecture is illustrated, and finally results from a fully implemented prototype are presented.
1