In this paper, translation pattern is evaluated based on its contribution to machine translation assessment score of training corpus.
论文以翻译模板对训练语料库机器译文评测分数的贡献为依据，对其进行评价。
In this pattern, each data source has its own proprietary semantic meaning, and semantic translation is performed in a point-to-point manner.
在此模式中，每个数据源都有其专有的语义，而且采用点对点的方式进行语义转换。
Translation knowledge unsuitable to target domain is deleted from pattern base, which leads MT system transfered to new domain quickly.
从模板库中过滤出适应目标领域的翻译知识，以实现系统向新领域的快速移植。
应用推荐