Speaker: Dr. Muyun YANG

Post Optimization Model for Machine Translation


Machine translation refers to the automatic translation of one natural language into another via computer, usually targeting at an optimal output once and for all. Referring to the manual process, however, this process is not completed in one time, and the checking, verification and formatting is involved before the final translation could be delivered. To address this phase which is less touched in the literature, this talk focuses on the post optimization of machine translation outputs via an additional translation memory. It presents two approaches for this purpose: an example based translation fusion model and a pseudo feedback based post-editing model. In contrast to the existing efforts in integrating the translation memory in the translation model, the proposed strategy emphasizes a system independent solution, which can be deployed with any third-party MT system.

Dr. Muyun YANG

About Dr. Muyun Yang

Muyun Yang is an Associate Professor in the School of computer Science and Technology at Harbin Institute of Technology (HIT). He received his Ph.D. in Computer Science from HIT in 2003 and his M.A. in Applied Linguistics in 1996. He is the member of NLP TC of CCF, and CL/IR/HealthBio TC of CIPSC. He has served on the editorial board of Journal of Chinese Information Process, and the PC of CWMT 2016, area (co-)chair of NLP-CC 2015, and several PC members of ACL and IJCAI. His research interests include machine translation, information retrieval and question answering system. His group ranks top in the automatic evaluation metric track of WMT 2010, as well as the micro-blog track in TREC 2013.