arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.01175cs.CL
收藏

语码转换数据的词汇规范化及其对词性标注的影响

Lexical Normalization for Code-switched Data and its Effect on POS-tagging

IT University of Copenhagen(哥本哈根信息技术大学);University of Stuttgart(斯图加特大学)

Rob van der Goot, Özlem Çetinoğlu

摘要:

词汇规范化,将非规范数据转换为标准语言,已被证明可以提高社交媒体上许多自然语言处理任务的性能。然而,在一个话语中使用多种语言,也称为代码转换(CS),经常被这些规范化系统所忽视,尽管它在社交媒体中很常见。在本文中,我们提出了三个规范化模型,专门设计用于处理代码转换的数据,我们评估两种语言对:土耳其语-英语(Id-En)和土耳其语-德语(Tr-De)。对于后者,我们为数据集引入了新的规范化层及其相应的语言ID和POS标签,并评估了规范化对POS标签的下游影响。结果表明,我们的CS定制的规范化模型优于Id-En国家的最先进的和Tr-De单语模型,并导致5.4%的相对性能提高的POS标记相比,非规范化的输入。

英文摘要:

Lexical normalization, the translation of non-canonical data to standard language, has shown to improve the performance of manynatural language processing tasks on social media. Yet, using multiple languages in one utterance, also called code-switching (CS), is frequently overlooked by these normalization systems, despite its common use in social media. In this paper, we propose three normalization models specifically designed to handle code-switched data which we evaluate for two language pairs: Indonesian-English (Id-En) and Turkish-German (Tr-De). For the latter, we introduce novel normalization layers and their corresponding language ID and POS tags for the dataset, and evaluate the downstream effect of normalization on POS tagging. Results show that our CS-tailored normalization models outperform Id-En state of the art and Tr-De monolingual models, and lead to 5.4% relative performance increase for POS tagging as compared to unnormalized input.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑