A Part-of-Speech Tagger for Yiddish

TL;DR

This paper presents the development and evaluation of a part-of-speech tagger for Yiddish, leveraging both historical and OCR'd text resources, and demonstrates improved performance through contextualized embeddings.

Contribution

It introduces a novel Yiddish POS tagger that combines historical and OCR'd corpora with contextualized embeddings, addressing orthographic inconsistencies.

Findings

Contextualized embeddings improve tagger accuracy.

Simple embeddings capture spelling variants without standardization.

Yiddish text resources enhance NLP tools for less-resourced languages.

Abstract

We describe the construction and evaluation of a part-of-speech tagger for Yiddish. This is the first step in a larger project of automatically assigning part-of-speech tags and syntactic structure to Yiddish text for purposes of linguistic research. We combine two resources for the current work - an 80K-word subset of the Penn Parsed Corpus of Historical Yiddish (PPCHY) and 650 million words of OCR'd Yiddish text from the Yiddish Book Center (YBC). Yiddish orthography in the YBC corpus has many spelling inconsistencies, and we present some evidence that even simple non-contextualized embeddings trained on YBC are able to capture the relationships among spelling variants without the need to first "standardize" the corpus. We also use YBC for continued pretraining of contexualized embeddings, which are then integrated into a tagger model trained and evaluated on the PPCHY. We evaluate the tagger performance on a 10-fold cross-validation split, showing that the use of the YBC text for the contextualized embeddings improves tagger performance. We conclude by discussing some next steps, including the need for additional annotated training and test data.