摘要
研究提高翻译算法的翻译准确性问题。由于英语名词短语语法复杂、规则多,导致传统机器翻译算法存在效率低、准确率不高难题。为提高英语名词短语识别准确率,提出一种粗糙集的名词短语识别方法。粗糙集方法把英语名词短语识别当成一个决策问题,通过粗糙集理论对特征约简和规则优化,最后对其进行进识别。采用粗糙集方法对WSJ英语名词短语样本进行仿真实验,仿真结果表明,粗糙集的名词短语识别正确率高于其它翻译方法,是一种有效英语名词短语机器识别方法。为实际设计提供依据。
English basenp recognition problem is Studied. Because English basenp grammar rules is very complex, the traditional English basenp recognition algorithm existing low efficiency and low accuracy problem. In order to improve basenp recognition accuracy, and put forward an English basenp recognition method based on rough set this method takes the English basenp recognition problem as a decision-making problem, by feature reduction and rules the features of rough set theory to solve. Simulation experiments are carried out on WSJ English basenp sample, the simulation results show that the recognition based on rough set theory method of phrases recognition rate is very high precision and recall, far superior other identification method, is a intelligence, accurate English basenp recognition method.
出处
《科技通报》
北大核心
2013年第10期26-29,共4页
Bulletin of Science and Technology
关键词
名词短语
机器翻译
粗糙集
basenp
machine translation
rough sets