If all you have is a bit of the Bible: Learning POS taggers for truly low-resource languages
If all you have is a bit of the Bible: Learning POS taggers for truly low-resource languages
复制标题
如果你只有一点圣经：学习真正的低资源语言的词性标注器
DOI：
10.3115/v1/p15-2044
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
Anders Søgaard
中科院分区：
文献类型：
--
作者：
Zeljko Agic;Dirk Hovy;Anders Søgaard
我们提出了一个简单的方法来学习像Akawaio，Aukan或Cakchiquel这样的语言的词性标记-这些语言只存在圣经部分的翻译。通过聚合来自几种注释语言的标签，并通过诗句上的单词对齐来传播它们，我们学习了100种语言的POS标签，使用这些语言相互引导。我们评估我们的跨语言模型的测试集存在的25种语言，以及另外10个，我们有标签字典。我们的方法比从圣经翻译中诱导的最先进的无监督POS标记器表现得更好（20-30%），并且通常与弱监督方法竞争，这些方法假设高质量的平行语料库，具有完美标记化的代表性单语语料库和/或标签词典。我们为所有100种语言提供模型。
We present a simple method for learning part-of-speech taggers for languages like Akawaio, Aukan, or Cakchiquel – languages for which nothing but a translation of parts of the Bible exists. By aggregating over the tags from a few annotated languages and spreading them via wordalignment on the verses, we learn POS taggers for 100 languages, using the languages to bootstrap each other. We evaluate our cross-lingual models on the 25 languages where test sets exist, as well as on another 10 for which we have tag dictionaries. Our approach performs much better (20-30%) than state-of-the-art unsupervised POS taggers induced from Bible translations, and is often competitive with weakly supervised approaches that assume high-quality parallel corpora, representative monolingual corpora with perfect tokenization, and/or tag dictionaries. We make models for all 100 languages available.