• DARIAH EU
• Publications close
• Other research products close
• EU close
• ES close
• HAL-Inserm close
• arXiv.org e-Print Archive close

In this paper we describe the development and evaluation of a visual analytics tool to support historical research. Historians continuously gather data related to their scholarly research from archival visits and background search. Organising and making sense of all this data can be challenging as many historians continue to rely on analog or basic digital tools. We built an integrated note-taking environment for historians which unifies a set of func-tionalities we identified as important for historical research including editing, tagging, searching, sharing and visualization. Our approach was to involve users from the initial stage of brainstorming and requirement analysis through to design, implementation and evaluation. We report on the process and results of our work, and conclude by reflecting on our own experience in conducting user-centered visual analytics design for digital humanities. International audience

En este trabajo presentamos un nuevo conjunto de datos y benchmark orientados a la tarea de similitud semántica en letras de canciones. Nuestro conjunto de datos, originalmente formado por 2775 pares de canciones en Español, fue anotado en un experimento de anotación colectivo por 63 anotadores nativos. Después de recoger y refinar los datos para asegurar un alto grado de consenso e integridad en los datos, obtuvimos 676 pares anotados de alta calidad que fueron empleados para evaluar el rendimiento de diferentes modelos del lenguaje monolingües y multilingües pertenecientes al estado del arte. En consecuencia, obtuvimos unos resultados base que esperamos sean de utilidad a la comunidad en todas aquellas aplicaciones académicas e industriales futuras que se realicen en este contexto. In this paper, we present a new dataset and benchmark tailored to the task of semantic similarity in song lyrics. Our dataset, originally consisting of 2775 pairs of Spanish songs, was annotated in a collective annotation experiment by 63 native annotators. After collecting and refining the data to ensure a high degree of consensus and data integrity, we obtained 676 high-quality annotated pairs that were used to evaluate the performance of various state-of-the-art monolingual and multilingual language models. Consequently, we established baseline results that we hope will be useful to the community in all future academic and industrial applications conducted in this context. This research has been carried out in the framework of the Grant LyrAIcs Grant agreement ID: 964009 funded by ERC-POCLS, and in the framework of the Grant CLS INFRA reference 101004984 funded by H2020-INFRAIA-2020-1. It has also received funding from the project ISL: Intelligent Systems for Learning (GID2016-39) in the call PID 22/23, and from FAIRTRANSNLP-DIAGNOSIS: Measuring and quantifying bias and fairness in NLP systems, grant PID2021-124361OB-C32, funded by MCIN/AEI/10.13039/501100011033 and by ERDF, EU A way of making Europe. Alejandro Benito-Santos acknowledges support from the postdoctoral grant ”Margarita Salas”, awarded by the Spanish Ministry of Universities.

This paper presents an attempt to provide a generic named-entity recognition and disambiguation module (NERD) called entity-fishing as a stable online service that demonstrates the possible delivery of sustainable technical services within DARIAH, the European digital research infrastructure for the arts and humanities. Deployed as part of the national infrastructure Huma-Num in France, this service provides an efficient state-of-the-art implementation coupled with standardised interfaces allowing an easy deployment on a variety of potential digital humanities contexts. Initially developed in the context of the FP9 EU project CENDARI, the software was well received by the user community and continued to be further developed within the H2020 HIRMEOS project where several open access publishers have integrated the service to their collections of published monographs as a means to enhance retrieval and access. entity-fishing implements entity extraction as well as disambiguation against Wikipedia and Wikidata entries. The service is accessible through a REST API which allows easier and seamless integration, language independent and stable convention and a widely used service-oriented architecture (SOA) design. Input and output data are carried out over a query data model with a defined structure providing flexibility to support the processing of partially annotated text or the repartition of text over several queries. The interface implements a variety of functionalities, like language recognition, sentence segmentation and modules for accessing and looking up concepts in the knowledge base. The API itself integrates more advanced contextual parametrisation or ranked outputs, allowing for the resilient integration in various possible use cases. The entity-fishing API has been used as a concrete use case to draft the experimental stand-off proposal, which has been submitted for integration into the TEI guidelines. The representation is also compliant with the Web Annotation Data Model (WADM). In this paper we aim at describing the functionalities of the service as a reference contribution to the subject of web-based NERD services. In this paper, we detail the workflow from input to output and unpack each building box in the processing flow. Besides, with a more academic approach, we provide a transversal schema of the different components taking into account non-functional requirements in order to facilitate the discovery of bottlenecks, hotspots and weaknesses. We also describe the underlying knowledge base, which is set up on the basis of Wikipedia and Wikidata content. We conclude the paper by presenting our solution for the service deployment: how and which the resources where allocated. The service has been in production since Q3 of 2017, and extensively used by the H2020 HIRMEOS partners during the integration with the publishing platforms. International audience

The CENDARI infrastructure is a research-supporting platform designed to provide tools for transnational historical research, focusing on two topics: medieval culture and World War I. It exposes to the end users modern Web-based tools relying on a sophisticated infrastructure to collect, enrich, annotate, and search through large document corpora. Supporting researchers in their daily work is a novel concern for infrastructures. We describe how we gathered requirements through multiple methods to understand historians' needs and derive an abstract workflow to support them. We then outline the tools that we have built, tying their technical descriptions to the user requirements. The main tools are the note-taking environment and its faceted search capabilities; the data integration platform including the Data API, supporting semantic enrichment through entity recognition; and the environment supporting the software development processes throughout the project to keep both technical partners and researchers in the loop. The outcomes are technical together with new resources developed and gathered, and the research workflow that has been described and documented. International audience

Biased language commonly occurs around topics which are of controversial nature, thus, stirring disagreement between the different involved parties of a discussion. This is due to the fact that for language and its use, specifically, the understanding and use of phrases, the stances are cohesive within the particular groups. However, such cohesiveness does not hold across groups. In collaborative environments or environments where impartial language is desired (e.g. Wikipedia, news media), statements and the language therein should represent equally the involved parties and be neutrally phrased. Biased language is introduced through the presence of inflammatory words or phrases, or statements that may be incorrect or one-sided, thus violating such consensus. In this work, we focus on the specific case of phrasing bias, which may be introduced through specific inflammatory words or phrases in a statement. For this purpose, we propose an approach that relies on a recurrent neural networks in order to capture the inter-dependencies between words in a phrase that introduced bias. We perform a thorough experimental evaluation, where we show the advantages of a neural based approach over competitors that rely on word lexicons and other hand-crafted features in detecting biased language. We are able to distinguish biased statements with a precision of P=0.92, thus significantly outperforming baseline models with an improvement of over 30%. Finally, we release the largest corpus of statements annotated for biased language. Comment: The Twelfth ACM International Conference on Web Search and Data Mining, February 11--15, 2019, Melbourne, VIC, Australia

Le quartier de Renancourt, situé à l’ouest de la ville d’Amiens, est connu dans la littérature archéologique depuis le début du XXe siècle par les travaux de V. Commont. Jusqu’à une date récente, ce gisement de plein air est resté l’un des rares témoignages du Paléolithique supérieur ancien pour l’ensemble de la région loessique du nord de la France. En 2011, une nouvelle concentration de vestiges, découverte à proximité immédiate des premières fouilles de V. Commont a été mise au jour lors d’un diagnostic archéologique lié à un projet d’aménagement. Ce gisement, désormais appelé Amiens-Renancourt 1, fait l’objet depuis 2014 de campagnes de fouilles programmées annuelles.L’occupation archéologique, située à 4 m de profondeur, est datée de 23000 BP, soit aux alentours de 27 000 ans cal. BP. Aux vestiges classiques et fréquents pour les gisements paléolithiques (industrie lithique, restes osseux, etc…) s’ajoute la découverte exceptionnelle de plusieurs statuettes féminines et d’éléments de parure en craie. La rareté et le nombre important (une quinzaine de statuettes à ce jour) de ces objets nous a incité à créer en 2016 un Projet Collectif de Recherche dédié, soutenu par le Service Régional de l’Archéologie des Hauts-de-France. Un des axes de ce projet a pour ambition la numérisation de l’ensemble des statuettes à des fins multiples : d’étude scientifique, de préservation, de muséographie et de valorisation.Le poster que nous soumettons vise à présenter les choix méthodologiques effectués pour mener à bien la numérisation, les difficultés techniques rencontrées dues notamment aux très petites dimensions des certaines statuettes et les résultats déjà obtenus et attendus. C’est in fine le cahier des charges du volet modélisation tri-dimensionnelle du projet qui sera présenté, incrémenté du retour d’expérience acquis lors des premières modélisations et impressions de Vénus, et des exploitations et des explorations actuelles et envisagées des modèles. National audience