Overview of ADoBo 2021: Automatic Detection of Unassimilated Borrowings in the Spanish Press

TL;DR

This paper reviews the ADoBo 2021 shared task on detecting English-origin lexical borrowings in Spanish news texts, highlighting the challenge and potential for NLP advancements.

Contribution

It presents the shared task setup, dataset, participant results, and insights into the difficulty of automatic borrowing detection in Spanish.

Findings

Results ranged from F1 scores of 37 to 85.

Detection is challenging with out-of-domain and OOV words.

Traditional lexicographic methods could improve with modern NLP techniques.

Abstract

This paper summarizes the main findings of the ADoBo 2021 shared task, proposed in the context of IberLef 2021. In this task, we invited participants to detect lexical borrowings (coming mostly from English) in Spanish newswire texts. This task was framed as a sequence classification problem using BIO encoding. We provided participants with an annotated corpus of lexical borrowings which we split into training, development and test splits. We received submissions from 4 teams with 9 different system runs overall. The results, which range from F1 scores of 37 to 85, suggest that this is a challenging task, especially when out-of-domain or OOV words are considered, and that traditional methods informed with lexicographic information would benefit from taking advantage of current NLP trends.