Detecting Unassimilated Borrowings in Spanish: An Annotated Corpus and Approaches to Modeling

TL;DR

This paper introduces a new annotated corpus of Spanish newswire with unassimilated borrowings and evaluates various sequence labeling models, finding that BiLSTM-CRF with subword embeddings and Transformer-based embeddings outperform multilingual BERT.

Contribution

It provides a large, detailed corpus for borrowing detection and compares multiple models, highlighting effective approaches for identifying unassimilated lexical borrowings in Spanish.

Findings

BiLSTM-CRF with subword embeddings performs best.

Transformer-based embeddings improve borrowing detection.

Corpus is larger and more diverse than previous resources.

Abstract

This work presents a new resource for borrowing identification and analyzes the performance and errors of several models on this task. We introduce a new annotated corpus of Spanish newswire rich in unassimilated lexical borrowings -- words from one language that are introduced into another without orthographic adaptation -- and use it to evaluate how several sequence labeling models (CRF, BiLSTM-CRF, and Transformer-based models) perform. The corpus contains 370,000 tokens and is larger, more borrowing-dense, OOV-rich, and topic-varied than previous corpora available for this task. Our results show that a BiLSTM-CRF model fed with subword embeddings along with either Transformer-based embeddings pretrained on codeswitched data or a combination of contextualized word embeddings outperforms results obtained by a multilingual BERT-based model.