As many MT users and developers know, building a domain-specific engine is not a goal in and of itself. It is just the first step in a long venture for that specific domain and language pair. Once the engine is built, it should be improved and optimized according to your needs in order to properly utilize the technology. In the meantime, if you are an LSP, you have to manage the production workflow. This webinar will cover the issues encountered during these processes using the language pair EN-TR as a case study. Topics include specific error typologies, post-editor’s involvement, gap analysis, and data patching.
Selçuk Özcan has 5+ years of experience in the language industry and is one of the founders of Transistent Language Automation Services. He holds double degree in Mechanical Engineering and Translation Studies. He has a keen interest in linguistics, NLP, language automation procedures, agile management, and technology integration. He is responsible for building high quality production models including QE and deploying “train the trainers” model. He also teaches Computer-aided Translation and Total Quality Management at Istanbul Yeni Yuzyil University, Translation & Interpreting Department.