Abstract: This paper presents Machine Translation(MT) model for extracting information and its application and problems in water information telegraph translation, which is based on rules of water information telegraph. It introduces semantic analysis technique to solve the problems, which is a kind of language processing technique in MT, and provides semantic information to increase the rate of translation. By applying logic semantics, it constructs the logic semantic framework of telegraph to obtain the logic semantics of telegraphs. Semantics-based water information translation model is presented, and its system is evaluated. The result shows the degree of automation is improved, and the telegraphs unable to be translated descend from 2%~5% to about 0.96%.

Key words: water information telegraph translation, logic semantics, semantic analysis

摘要： 介绍针对信息提取的机器翻译模型，分析该模型下基于语法分析的水情电报翻译系统及存在的问题。采用机器翻译中语义分析技术解决问题，利用语义信息提高翻译率。用逻辑语义建立水情电报的逻辑语义模型，结合语义信息实现对几种类型错报的翻译，由此提出基于语义的水情电报翻译模型。对系统的评价与分析结果表明，翻译率即自动化程度得到了提高，报文的不可翻译率从2%~5%降到0.96%左右。

关键词: 水情电报翻译, 逻辑语义, 语义分析

CLC Number: 

• N945