In recent years the availability of medieval charter texts has increased thanks to advances in OCR and HTR techniques. But the lack of models that automatically structure the textual output continues to hinder the extraction of large-scale lectures from these historical sources that are among the most important for medieval studies. This paper presents the process of annotating and modelling a corpus to automatically detect named entities in medieval charters in Latin, French and Spanish and address the problem of multilingual writing practices in the Late Middle Ages. It introduces a new annotated multilingual corpus and presents a training pipeline using two approaches: (1) a method using contextual and static embeddings coupled to a Bi-LSTM-CRF classifier; (2) a fine-tuning method using the pre-trained multilingual BERT and RoBERTa models. The experiments described here are based on a corpus encompassing about 2.3M words (7576 charters) coming from five charter collections ranging from the 10th to the 15th centuries. The evaluation proves that both multilingual classifiers based on general purpose models and those specifically designed achieve high-performance results and do not show performance drop compared to their monolingual counterparts. This paper describes the corpus and the annotation guideline, and discusses the issues related to the linguistic of the charters, the multilingual writing practices, so as to interpret the results within a larger historical perspective.

This paper presents the process of annotating and modelling a corpus to automatically detect named entities in medieval charters in French. It introduces a new annotated corpus and a new system which outperforms state-of-the art libraries. Charters are legal documents and among the most important historical sources for medieval studies as they reflect economic and social dynamics as well as the evolution of literacy and writing practices. Automatic detection of named entities greatly improves the access to these unstructured texts and facilitates historical research. The experiments described here are based on a corpus encompassing about 500k words (1200 charters) coming from three charter collections of the 13th and 14th centuries. We annotated the corpus and then trained two state-of-the art NLP libraries for Named Entity Recognition (Spacy and Flair) and a custom neural model (Bi-LSTM-CRF). The evaluation shows that all three models achieve a high performance rate on the test set and a high generalization capacity against two external corpora unseen during training. This paper describes the corpus and the annotation model, and discusses the issues related to the linguistic processing of medieval French and formulaic discourse, so as to interpret the results within a larger historical perspective.