Statistical inflectional processing of agglutinative languages

粘着语言的统计屈折处理

• 批准号: 22700143
• 负责人: OGAWA Yasuhiro
• 金额: $ 2.25万
• 依托单位: Nagoya University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Young Scientists (B)
• 财政年份: 2010
• 资助国家: 日本
• 起止时间: 2010 至 2011
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-22700143/
• 关键词: 膠着語; ウイグル語; ウズベク語; 形態素解析; 音韻変化

Japanese, Uighur and Uzbek languages are agglutinative and they have various types of inflection. Traditional ways to deal with the inflection are rule-based methods. In contrast to this, we considered inflection is a kind of transliteration and introduced a statistical approach. Thus we tried automatic generation of inflection rules based on statistical machine translation techniques. From this, we obtained the results showing the effectiveness of statistical inflectional processing.

日语、维吾尔语和乌孜别克语都是黏着语，它们有各种各样的屈折变化。处理拐点的传统方法是基于规则的方法。与此相反，我们认为屈折变化是一种音译，并介绍了一种统计方法。因此，我们尝试了基于统计机器翻译技术的自动生成屈折变化规则。由此，我们得到的结果表明统计屈折处理的有效性。