摘要
近年来,机器翻译利用人工智能技术和专业人工翻译技术的相互耦合,能够满足对跨语种语言交流需求。目前机器翻译的产品服务已大规模应用于国际贸易、旅游、教育和日常生活当中,如何统一评估相关产品服务的质量成为亟待研究和解决的问题。基于国内外人工智能和机器翻译的标准分析,构建机器翻译系统的通用模型,建立译文质量和系统能力两个维度的机器翻译评估模型,在此基础上,开展机器翻译能力等级评估标准化并已应用于机器翻译评测验证。
In recent years,machine translation uses the coupling of artificial intelligence technology and human translation to meet the needs of cross language communication.At present,the product service of machine learning has been widely used in international trade,tourism,education and daily life.How to assess the quality of related products and services has become an urgent problem.Based on the standard analysis of artificial intelligence and machine translation,this paper constructs the general model of machine translation system,establishes the assessment model of machine translation with two dimensions of translation quality and system capability.It carries out the standardization of machine translation capability classified assessment and has been applied to the testing and verification of machine translation.
出处
《信息技术与标准化》
2020年第1期21-26,共6页
Information Technology & Standardization
基金
教育部人文社会科学研究青年基金项目“基于关联数据语义扩展的标准知识语义链接服务研究”,项目编号:17YJC870013
中国博士后科学基金项目“面向标准信息服务的领域标准知识语义链接研究”,项目编号:2017M622527
湖北省自然科学基金项目“基于关联数据的标准知识挖掘与信息服务关键技术研究”,项目编号:2018CFB181。
关键词
人工智能
机器翻译
评估模型
标准化
artificial intelligence
machine translation
assessment model
standardization