Introducing the HIPE 2022 Shared Task: Named Entity Recognition and Linking in Multilingual Historical Documents

Ehrmann, Maud; Romanello, Matteo; Doucet, Antoine; Clematide, Simon

doi:10.1007/978-3-030-99739-7_44

Cited by 7 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The StanfordNER-Toolkit is frequently used to train or reuse genre-specific classifiers to annotate historical documents such as letters (cf. Ehrmann et al 2021) and can be considered as a well-established tool for NER tasks. The software uses Conditional Random Field (CRF) algorithms (cf.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Geographical Research in the Digital Humanities

2024

Digital Humanities Research

View full text Add to dashboard Cite

The richness of social and cultural theory in the humanities offers countless opportunities for using theory-informed concepts in data-based analysis workflows. The contributors to this volume thus encourage further research utilizing out-of-the-box models and approaches to space and place in the field of Digital Humanities. The collection follows the two complementary goals of providing promising conceptualisations of space and place for a broad audience from Digital Humanities, and of presenting current work in Digital Humanities using different conceptualisations of space and place or offering innovative methods for their analysis.

show abstract

Section: Methodsmentioning

confidence: 99%

“…An automated and transferable method for extracting specific entities is already under discussion (cf. Hildenbrandt/Kamzelak 2019; Ehrmann et al 2021). A practical application has so far remained elusive but is currently developed in the course of the project Dehmel digital (cf.…”

Section: Modelling Narrative Space In Novels and Letter Correspondencesmentioning

confidence: 99%

Geographical Research in the Digital Humanities

2024

Digital Humanities Research

View full text Add to dashboard Cite

show abstract

“…[5]. The development of various neural-based approaches helped increase the performance of NER systems on historical materials with F-scores going from 60-70% on average for rule based and traditional ML systems to, for the best neural systems, 80% [6].…”

Section: Analysis Of Research and Publicationsmentioning

confidence: 99%

Adaptive Domain-Specific Named Entity Recognition Method With Limited Data

DYCHKA,

POTAPOVA,

VOVK

et al. 2024

ВИМІРЮВАЛЬНА ТА ОБЧИСЛЮВАЛЬНА ТЕХНІКА В ТЕХНОЛОГІЧНИХ ПРОЦЕСАХ

View full text Add to dashboard Cite

The ever-evolving volume of digital information requires the development of innovative search strategies aimed at obtaining the necessary data efficiently and economically feasible. The urgency of the problem is emphasized by the growing complexity of information landscapes and the need for fast data extraction methodologies. In the field of natural language processing, named entity recognition (NER) is an essential task for extracting useful information from unstructured text input for further classification into predefined categories. Nevertheless, conventional methods frequently encounter difficulties when confronted with a limited amount of labeled data, posing challenges in real-world scenarios where obtaining substantial annotated datasets is problematic or costly. In order to address the problem of domain-specific NER with limited data, this work investigates NER techniques that can overcome these constraints by continuously learning from newly collected information on pre-trained models. Several techniques are also used for making the greatest use of the limited labeled data, such as using active learning, exploiting unlabeled data, and integrating domain knowledge. Using domain-specific datasets with different levels of annotation scarcity, the fine-tuning process of pre-trained models, such as transformer-based models (TRF) and Toc2Vec (token-to-vector) models is investigated. The results show that, in general, expanding the volume of training data enhances most models' performance for NER, particularly for models with sufficient learning ability. Depending on the model architecture and the complexity of the entity label being learned, the effect of more data on the model's performance can change. After increasing the training data by 20%, the LT2V model shows the most balanced growth in accuracy overall by 11% recognizing 73% of entities and processing speed. Meanwhile, with consistent processing speed and the greatest F1-score, the Transformer-based model (TRF) shows promise for effective learning with less data, achieving 74% successful prediction and a 7% increase in performance after expanding the training data to 81%. Our results pave the way for the creation of more resilient and efficient NER systems suited to specialized domains and further the field of domain-specific NER with sparse data. We also shed light on the relative merits of various NER models and training strategies, and offer perspectives for future research.

show abstract

“…The digitization of newspapers has greatly improved accessibility and clearly changed the nature of historical research, by enabling easier data access and analysis at scale through multilingual semantic data enrichment [7,42,10,6]. Through better document analysis results and semantic enrichment e.g., named entity recognition (NER), relation extraction (RE), event extraction (EE), the quality of the newspaper data offered by the libraries to its users is substantially improved [12,14,13,15]. Preserving the historical memory of entities and events from historical documents and making them accessible to a larger audience, not only limited to humanities scholars and experts, could lead to better organization of our historical knowledge [44,2,38].…”

Section: Introductionmentioning

confidence: 99%

Experimenting with Unsupervised Multilingual Event Detection in Historical Newspapers

Boroş

Cabrera-Diego

Doucet

2022

From Born-Physical to Born-Virtual: Augmenting Intelligence in Digital Libraries

Self Cite

View full text Add to dashboard Cite

To prevent historical knowledge's fading, research in event detection could facilitate access to digitized collections. In this paper, we propose a method for annotating multilingual historical documents for event detection in an unsupervised manner by leveraging entities and semantic notions of event types. We automatically annotate the documents by relying on dependency parse trees and automatic semantic mapping to event-based frames, with a focus on the multilingual transfer between frames and candidate events. The documents are afterward verified by native speakers, Digital Humanities researchers. We also report on experimental results of event detection in historical newspapers with a state-of-the-art model. We demonstrate that our approach allows for easy language adaptation by presenting two study cases with knowledge extracted from German newspapers from 1911 to 1933 regarding events surrounding International Women's Day and from French newspapers between 1900 and 1944 related to the abolition of guillotine executions in France. Our preliminary findings show that this type of approach could alleviate the need for manual annotation by also providing a practical course of action toward unsupervised event detection from multilingual digitized and historical documents.

show abstract

Introducing the HIPE 2022 Shared Task: Named Entity Recognition and Linking in Multilingual Historical Documents

Cited by 7 publications

References 11 publications

Geographical Research in the Digital Humanities

Geographical Research in the Digital Humanities

Adaptive Domain-Specific Named Entity Recognition Method With Limited Data

Experimenting with Unsupervised Multilingual Event Detection in Historical Newspapers

Contact Info

Product

Resources

About