基于词向量空间模型的机器翻译质量评价分析——以石油术语有道翻译为例

The Results Analysis of Machine Translation on Petroleum Engineering Terms Based on Vector Space Model

文章从问题意识视角出发,以石油术语为基础,引入词向量空间模型的方法展开三个相关实验对机器译文和人工译文进行对比研究,探索机器翻译结果在空间模型中的演绎和呈现。实验结果显示机器翻译对于石油术语的语言翻译准度能达到0.403。文章尝试结合计算机技术、语言学和翻译学等不同领域量化论证两种翻译结果在语义层面的接近和靠拢程度,以期探索评价分析机器翻译系统输出结果质量的新途径。

From the perspective of problem awareness,this paper conducted an in-depth terminology analysis on machine translation and manual translation by training vector space model.Three experiments were performed by the means of training the vector space model to compare the results of machine translation.These experiments demonstrate the similarity between machine translation and manual translation is 0.403.Integrated with computer technology,linguistics and translation,this paper focuses on the semantic similarity between machine translation and manual translation that aims to blaze a new way for results evaluation of machine translation.