海量数据机器单词中关键语义筛选方法研究

发布时间：2018-05-15 21:08

  本文选题：海量数据 + 机器翻译　； 参考：《现代电子技术》2017年06期

【摘要】：为了提高机器单词翻译的准确性,需要进行关键语义筛选和特征提取,故提出一种基于主题词表自然语义信息抽取的海量数据机器单词中关键语义筛选方法。首先构建海量数据机器单词的文本语义主题词概念决策树模型,采用语义信息转换方式计算机器单词中关键语义的利用规则、聚类中心等信息参量;然后采用主题词表自然语义信息抽取方法进行语义评估和翻译可靠性测试,实现关键语义自动筛选控制;最后进行仿真测试。结果表明,采用该方法进行机器单词中关键语义筛选,提高了文本机器翻译的自适应配准能力,翻译的准确性得到有效提高。
[Abstract]:In order to improve the accuracy of machine word translation, it is necessary to carry out the key semantic filtering and feature extraction. Therefore, a method of key semantic selection in massive data machine words based on the natural semantic information extraction of topic vocabulary is proposed. Firstly, the conceptual decision tree model of text semantic subject words is constructed, and the key semantic usage rules, clustering centers and other information parameters in machine words are calculated by semantic information conversion. Then the semantic evaluation and translation reliability test are carried out by extracting the natural semantic information of the subject list to realize the automatic selection control of key semantics. Finally, the simulation test is carried out. The results show that this method can improve the adaptive registration ability of text machine translation and improve the accuracy of translation.
【作者单位】： 石河子大学;
【分类号】：H085
，

本文编号：1893937