Multilingual Event Extraction from Historical Newspaper Adverts

TL;DR

This paper introduces a multilingual dataset for event extraction from historical newspaper ads, demonstrating that extractive QA models and machine translation can effectively address low-resource challenges in historical NLP tasks.

Contribution

It presents a new multilingual dataset for historical event extraction and explores effective methods like extractive QA and machine translation for low-resource historical languages.

Findings

Extractive QA models perform well with scarce data.

Machine translation often outperforms low-resource learning methods.

Cross-lingual transfer remains highly challenging.

Abstract

NLP methods can aid historians in analyzing textual materials in greater volumes than manually feasible. Developing such methods poses substantial challenges though. First, acquiring large, annotated historical datasets is difficult, as only domain experts can reliably label them. Second, most available off-the-shelf NLP models are trained on modern language texts, rendering them significantly less effective when applied to historical corpora. This is particularly problematic for less well studied tasks, and for languages other than English. This paper addresses these challenges while focusing on the under-explored task of event extraction from a novel domain of historical texts. We introduce a new multilingual dataset in English, French, and Dutch composed of newspaper ads from the early modern colonial period reporting on enslaved people who liberated themselves from enslavement. We find that: 1) even with scarce annotated data, it is possible to achieve surprisingly good results by formulating the problem as an extractive QA task and leveraging existing datasets and models for modern languages; and 2) cross-lingual low-resource learning for historical languages is highly challenging, and machine translation of the historical datasets to the considered target languages is, in practice, often the best-performing solution.