Korean Language Support
To kick off 2021 we're pleased to announce support for a new language: Korean. Our full Entity Extraction, Topic Tagging and Classification pipeline is now available in Korean through our public API. Like all of our languages you don't need to do anything to try it out, just start sending Korean to our main analysis endpoint, or paste a document into our demo.
TextRazor doesn't rely on translation for its language support, for the best results we build independent models that can pick up the intricacies of each language. Each of these models is built to leverage our large knowledgebase of real life entity relationships, and is integrated in our daily incremental entity indexing process. You therefore shouldn't see much of a quality gap between Korean and our other languages, but if you do spot any teething problems please let us know.
Tweet
