Maud Ehrmann
EPFL CDH DHI DHLAB
INN 116 (Bâtiment INN)
Station 14
1015 Lausanne
Web site: Web site: https://dhlab.epfl.ch
+41 21 693 19 31
EPFL
>
CDH
>
CDH-SHS
>
SHS-ENS
+41 21 693 19 31
EPFL
>
CDH
>
CDH-SG
>
CDH-IT
Fields of expertise
With a background in both natural language processing (NLP) and the humanities, my expertise is in the area of historical and multilingual NLP, with particular focus on historical document processing, information extraction, named entity processing, multilingual and historical resource creation, NLP system evaluation, and large-scale infrastructure. In recent years, I have worked and coordinated work on these topics in research projects at the intersection of computer science and cultural heritage - an interdisciplinary setting in which I have often acted as an intermediary between computer scientists, humanities scholars, engineers, and representatives of cultural heritage institutions.
Highlights:
impresso. Media Monitoring of the Past. How can newspaper archives help understand the past? How to explore them? This large-scale, impact-driven project aims to enable critical mining of newspaper archives by integrating robust content mining and innovative data visualisation and exploration into a powerful user interface that can support digital scholarship.
The HIPE Evaluation Campaigns. What is the ability of machines to recognise and disambiguate entities (e.g. people, places, organisations) in multilingual historical documents? The series of HIPE shared tasks aims to assess and advance the development of robust, adaptable and transferable approaches to named entity processing in historical documents to foster efficient semantic indexing of digitised cultural heritage collections. See the HIPE-2020 and HIPE-2022 websites, the HIPE-eval GitHub organisation, the HIPE-2022 dataset, and the DHLAB web page.
Highlights:
impresso. Media Monitoring of the Past. How can newspaper archives help understand the past? How to explore them? This large-scale, impact-driven project aims to enable critical mining of newspaper archives by integrating robust content mining and innovative data visualisation and exploration into a powerful user interface that can support digital scholarship.
The HIPE Evaluation Campaigns. What is the ability of machines to recognise and disambiguate entities (e.g. people, places, organisations) in multilingual historical documents? The series of HIPE shared tasks aims to assess and advance the development of robust, adaptable and transferable approaches to named entity processing in historical documents to foster efficient semantic indexing of digitised cultural heritage collections. See the HIPE-2020 and HIPE-2022 websites, the HIPE-eval GitHub organisation, the HIPE-2022 dataset, and the DHLAB web page.
Biography
Maud Ehrmann is a research scientist and lecturer at the Digital Humanities Laboratory of the Ecole Polytechnique Fédérale de Lausanne. She holds a PhD in Computational Linguistics from the Paris Diderot Universtiy (Paris 7) and has been engaged in a large number of scientific projects centred on information extraction and text analysis, both for present-time and historical documents. Before joining the DHLAB, she worked at the Linguistics Computing Laboratory at the Sapienza University of Rome where she worked on the BabelNet resource and contributed to the LIDER project (2013-2014). Prior to that, she worked at the European Commission’s Joint Research Centre in Ispra, Italy, as member of the OPTIMA unit (now Text and Data mining unit) which develops innovative and application-oriented solutions (Europe Media Monitor) for retrieving and extracting information from the Internet with a focus on high multilinguality (2009-2013). Previously, she worked at the Xerox Europe Research Centre in Grenoble, France (now Naver Labs Europe) in the Parsing and Semantics unit, first as PhD candidate supported by a CIFRE grant (2005-2008), then as a post-doctoral researcher (2008-2009). There, her research focused mainly on the automatic processing and fine-grained analysis of entities of interest, specifically named entities and temporal expressions.Publications
Infoscience publications
Towards Chapterisation of Podcasts Detection of Host and Structuring Questions in Radio Transcripts
2024-03-28Advisor(s) : M. Ehrmann
Post-correction of Historical Text Transcripts with Large Language Models: An Exploratory Study
Proceedings of the 8th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (LaTeCH-CLfL 2024). 2024-02-18. The 8th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature , St Julian's, Malta , March 22, 2024. p. 133-159.impresso Text Reuse at Scale. An interface for the exploration of text reuse data in semantically enriched historical newspapers
Frontiers in Big Data
2023-11-03
Vol. 6, num. Visualizing Big Culture and History Data .DOI : 10.3389/fdata.2023.1249469
Where Did the News Come From? Detection of News Agency Releases in Historical Newspapers
2023-08-18Advisor(s) : M. Ehrmann; E. Boros; M. Duering; F. Kaplan
From Archival Sources to Structured Historical Information: Annotating and Exploring the "Accordi dei Garzoni"
Apprenticeship, Work, Society in Early Modern Venice; Abingdon: Routledge, Taylor & Francis Group,2023-02-10.
p. 304.DOI : 10.4324/9781003197195-6.
Computational Approaches to Digitised Historical Newspapers (Dagstuhl Seminar 22292)
2023
Overview of HIPE-2022: Named Entity Recognition and Linking in Multilingual Historical Documents
Experimental IR Meets Multilinguality, Multimodality, and Interaction. 13th International Conference of the CLEF Association, CLEF 2022, Bologna, Italy, September 5–8, 2022, Proceedings. 2022. 13th Conference and Labs of the Evaluation Forum (CLEF 2022) , Bologna, Italy , 5-8 September 2022. p. 423-446.DOI : 10.1007/978-3-031-13643-6_26.
Digitised Historical Newspapers: A Changing Research Landscape (Introduction)
Digitised Newspapers – A New Eldorado for Historians?; Berlin, Boston: De Gruyter Oldenbourg,2022-12-31.
p. 439.
Digitised Newspapers – A New Eldorado for Historians? Reflections on Tools, Methods and Epistemology
Berlin: De Gruyter.Extended Overview of HIPE-2022: Named Entity Recognition and Linking in Multilingual Historical Documents
Proceedings of the Working Notes of CLEF 2022 - Conference and Labs of the Evaluation Forum. 2022. 13th Conference and Labs of the Evaluation Forum (CLEF 2022) , Bologna, Italy , 5-8 Sept 2022.DOI : 10.5281/zenodo.6979577.
ECCE: Entity-centric Corpus Exploration Using Contextual Implicit Networks
WWW ’22 Companion. 2022. The Web Conference (WWW'22) , Lyon, France , April 25-29, 2022. p. 1-4.DOI : 10.1145/3487553.3524237.
Introducing the HIPE 2022 Shared Task: Named Entity Recognition and Linking in Multilingual Historical Documents
Advances in Information Retrieval. 2022-04-05. 44th European Conference on IR Research, ECIR 2022 , Stavanger, Norway , April 10-14, 2022. p. 347-354.DOI : 10.1007/978-3-030-99739-7_44.
HIPE-2022 Shared Task Named Entity Datasets
2022.
Automatic table detection and classification in large-scale newspaper archives
2022-02-08Advisor(s) : M. Ehrmann; S. Clematide; F. Kaplan
Named Entity Recognition and Classification in Historical Documents: A Survey
ACM Computing Survey
2021-09-21
Vol. 56, num. 2.
Explorer la presse numérisée : le projet Impresso
Revue Historique Vaudoise
2021-11-27
Vol. 129/2021.Datasets and Models for Historical Newspaper Article Segmentation
2021.
Combining Visual and Textual Features for Semantic Segmentation of Historical Newspapers
Journal of Data Mining & Digital Humanities
2021
Vol. 2021, num. Special Issue on HistoInformatics: Computational Approaches to History.DOI : 10.5281/zenodo.4065271
CLEF-HIPE-2020 Shared Task Named Entity Datasets
2020.
Historical Newspaper Content Mining: Revisiting the impresso Project's Challenges in Text and Image Processing, Design and Historical Scholarship
DH2020 Book of Abstracts. 2020. Digital Humanities Conference (DH) , Ottawa, Canada , July 20-24, 2020.DOI : 10.5281/zenodo.4641894.
The impresso system architecture in a nutshell
2020
Impresso Named Entity Annotation Guidelines (CLEF-HIPE-2020)
2020
CLEF-HIPE-2020 - Shared Task Participation Guidelines
2020
Extended Overview of CLEF HIPE 2020: Named Entity Processing on Historical Newspapers
CLEF 2020 Working Notes. Conference and Labs of the Evaluation Forum. 2020-10-21. 11th Conference and Labs of the Evaluation Forum (CLEF 2020) , [Online event] , 22-25 September, 2020.DOI : 10.5281/zenodo.4117566.
Language Resources for Historical Newspapers: the Impresso Collection
Proceedings of the 12th Language Resources and Evaluation Conference. 2020-05-11. 12th International Conference on Language Resources and Evaluation (LREC) , Marseille, France , May 11-16 2020. p. 958-968.DOI : 10.5281/zenodo.4641902.
Overview of CLEF HIPE 2020: Named Entity Recognition and Linking on Historical Newspapers
Experimental IR meets multilinguality, multimodality, and interaction. 11th International Conference of the CLEF Association, CLEF 2020, Thessaloniki, Greece, September 22–25, 2020, Proceedings. 2020-09-15. 11th International Conference of the CLEF Association - CLEF 2020 , Thessaloniki, Greece , September 22–25, 2020. p. 288--310.DOI : 10.1007/978-3-030-58219-7_21.
Introducing the CLEF 2020 HIPE Shared Task: Named Entity Recognition and Linking on Historical Newspapers
Advances in Information Retrieval. ECIR 2020. 2020-04-08. ECIR 2020 : 42nd European Conference on Information Retrieval , Lisbon, Portugal , April 14-17, 2020. p. 524-532.DOI : 10.1007/978-3-030-45442-5_68.
Survey of digitized newspaper interfaces (dataset and notebooks)
2019.
The Past, Present and Future of Digital Scholarship with Newspaper Collections
DH 2019 Book of Abstracts. 2019-07-09. DIgital Humanities Conference , Utrecht , July 2019.Historical newspaper semantic segmentation using visual and textual features
2019-06-21Advisor(s) : M. Ehrmann; S. Ares Oliveira; S. Clematide
Beyond Keyword Search: Semantic Indexing and Exploration of Large Collections of Historical Newspapers
Digital Humanitites in the Nordic Countries, Copenhagen, Denmark, March 2019.Index-Driven Digitization and Indexation of Historical Archives
Frontiers in Digital Humanities
2019-03-11
Vol. 6, num. 1-16.DOI : 10.3389/fdigh.2019.00004
Historical Newspaper User Interfaces: A Review
[Proceedings of the 85th IFLA General Conference and Assembly]. 2019-09-02. 85th IFLA General Conference and Assembly , Athens, Greece , 24-30 August 2019. p. 1-24.DOI : 10.5281/zenodo.3404155.
Named Entity Processing for Historical Texts
2019-07-17.
Linked Lexical Knowledge Bases Foundations and Applications
Computational Linguistics
2017
Vol. 43, num. 2.DOI : 10.1162/COLI_r_00289
JRC-Names: Multilingual Entity Name variants and titles as Linked Data
Semantic Web
2017-01-02
Vol. 8, num. 2.DOI : 10.3233/SW-160228
From Documents to Structured Data: First Milestones of the Garzoni Project
DHCommons
2016
num. 2.Diachronic Evaluation of NER Systems on Old Newspapers
Proceedings of the 13th Conference on Natural Language Processing (KONVENS 2016). 2016. 13th Conference on Natural Language Processing (KONVENS 2016)Conference on Natural Language Processing , Bochum, GermanyBochum, Germany , September 19-21, 2016September 19–21, 2016. p. 97-107.Navigating through 200 years of historical newspapers
2016. International Conference on Digital Preservation (IPRES) , Bern, Switzerland , October 3-6, 2016.Cross-lingual Linking of Multi-word Entities and their corresponding Acronyms
Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016). 2016. 10th International Conference on Language Resources and Evaluation , Portorož, Slovenia , May 2016.Named Entity Resources - Overview and Outlook
Proceedings of the 9th International Conference on Language Resources and Evaluation. 2016. 10th International Conference on Language Resources and Evaluation , Portorož, Slovenia , May 2016.A Method for Record Linkage with Sparse Historical Data
2016. Digital Humanities Conference 2016 , Krakow, Poland , July 11-16, 2016.Les entités nommées pour le traitement automatique des langues
ISTE editions.Selected publications
| Maud Ehrmann, Ahmed Hamdi, Elvys Linhares Pontes, Matteo Romanello, Antoine Doucet. ACM Computing Survey (accepted) |
Named Entity Recognition and Classification in Historical Documents: A Survey |
| Maud Ehrmann, Matteo Romanello, Sven Najem-Meyer, Antoine Doucet, Simon Clematide. CLEF 2022 proceedings |
Extended Overview of HIPE-2022: Named Entity Recognition and Linking in Multilingual Historical Documents |
| Maud Ehrmann, Matteo Romanello, Alex Flückiger, Simon Clematide. CLEF 2020 proceedings |
Extended Overview of CLEF HIPE 2020: Named Entity Processing on Historical Newspapers |
| Maud Ehrmann, Matteo Romanello, Simon Clematide, Phillip Benjamin Ströbel, Raphaël Barman LREC 2020 |
Language Resources for Historical Newspapers: the Impresso Collection |
Teaching & PhD
Teaching
Humanities and Social Sciences Program
Courses
Digital humanities
Digital Humanities is a discipline at the crossroads of the information sciences and the humanities and social sciences. In this course, students discover this field of research by learning how to extract information from a large collection of historical newspapers.
Press and digital history I
At the crossroads of digital history, media and public history, this course focuses on the production, dissemination and conservation of information. Students will learn to take a critical look at the evolution of the media sphere through time and will be introduced to scientific mediation.
Press and digital history II
At the crossroads of digital history, media and public history, this course focuses on the production, dissemination and conservation of information. Students will learn to take a critical look at the evolution of the media sphere through time and will be introduced to scientific mediation.