首页> 外文会议>Asian Language Processing, 2009. IALP '09 >Identification of Closely Related Indigenous Languages: An Orthographic Approach
【24h】

Identification of Closely Related Indigenous Languages: An Orthographic Approach

机译:密切相关的土著语言的识别:正交法

获取原文

摘要

The main focus of this study is to identify the closely related languages amongst the indigenous languages of Sarawak and major languages such as Bahasa Melayu and English. The indigenous languages involved in this study include Iban (standard), Bidayuh (Bau-Jagoi), Kelabit (Bario), Melanau (Matu-Daro), Saȁ9;ban (Long Banga) and Penan (East Baram). The relationship between the languages is established via the proportion of cognates in the Swadesh list of the language pairs. The orthographic approach, which primarily examines the spelling of the vocabulary words, is used. The outcome of this study reveals that some indigenous languages are more closely related to to Bahasa Melayu than others. The findings from this research serve as an initial solution to answer the greater challenges in computational linguistics, such as the use of closely related languages as Pivot solutions in problems related to the under-resourced languages.
机译:这项研究的主要重点是在砂拉越的土著语言和主要的语言(如马来语和英语)中确定紧密相关的语言。参与这项研究的土著语言包括伊班语(标准),比达尤语(鲍-加戈伊),克拉比语(巴里奥),梅拉瑙(马图-达罗),Saȁ9;班(隆·班加)和佩南(东巴拉姆)。语言之间的关系是通过在语言对的Swadesh列表中的认知比例来建立的。使用主要检查词汇单词拼写的拼字法。这项研究的结果表明,某些土著语言与马来语的关系比其他语言更为紧密。这项研究的发现是应对计算语言学更大挑战的初步解决方案,例如在资源贫乏的语言相关问题中使用紧密相关的语言作为Pivot解决方案。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号