This paper proposes a two-phase example-based machine translation methodology which develops translation templates from examples and then translates using template matching. This ...
Traditional ranking mainly focuses on one type of data source, and effective modeling still relies on a sufficiently large number of labeled or supervised examples. However, in m...
Bo Wang, Jie Tang, Wei Fan, Songcan Chen, Zi Yang,...
Recently, models based on conditional random fields (CRF) have produced promising results on labeling sequential data in several scientific fields. However, in the vision task of c...
Huazhong Ning, Wei Xu, Yihong Gong, Thomas S. Huan...
Producing machine translation (MT) for the many minority languages in the world is a serious challenge. Minority languages typically have few resources for building MT systems. Fo...