A Comparison Study of Parsers for Patent Machine Translation

Machine translation of patent documents is very important from a practical point of view. One of the key technologies for improving machine translation quality is the utilization of syntax. It is difficult to select the appropriate parser for patent translation because the effects of each parser on patent translation are not clear. This paper provides comparative evaluation of several state-of-the-art parsers for English, focusing on the effects for patent machine translation from English to Japanese. We measured how much each parser contributed to improve translation quality when the parser was used to obtain the syntax of input sentences. In addition, we examined the effects of a method using parsed documentlevel context containing the input sentence to determine noun phrases (Onishi et al., 2011). We conducted experiments using the NTCIR8 patent translation task dataset. Most of the parsers improved translation quality. When the method using document-level context was applied, all of the compared parsers improved translation quality.