摘要
以高精度翻译多种自然语言的单词/语句为目标,设计基于人工智能的机器自动翻译系统。首先设计了机器自动翻译系统的总体结构,然后重点描述了机器自动翻译系统的核心模块,该模块获取单词/语句通过训练获取词向量,初始词向量并赋予词性特征,采用对数线性模型实现多种自然语言的单词/语句的词向量多特征融合翻译,最后进行了实例分析与验证。结果表明,该系统可高精度翻译自然语言,在不同句型、不同并发用户量、未登录词不同字符数量下,系统翻译性能较为稳定,能够满足实际应用需求。
In order to translate the words/sentences of many natural languages with high precision,a machine automatic translation system based on artificial intelligence is designed.The overall structure of the MT system is designed firstly,and then the core module of the MT system is described.The module obtains the word vector and the initial word vector,and gives the part of speech features through training.The log linear model is used to realize the word vector multi feature fusion translation of multiple natural languages.Finally,an example is analyzed.The results show that the system can translate natural languages with high accuracy.Under different sentence patterns,different concurrent users and different characters of non-login words,the system has stable translation performance and can meet the practical application requirements.
作者
霍小静
HUO Xiaojing(College of Humanities and International Education, Xi’an Peihua University, Xi’an 710125, China)
出处
《微型电脑应用》
2020年第11期77-79,共3页
Microcomputer Applications
基金
西安培华学院课堂教学模式创新专项项目(PHJM10904)。
关键词
人工智能
机器自动翻译
词向量
低频词
artificial intelligence
machine automatic translation
word vector
low frequency word
