• 2018-2022close
• Open Access close
• Part of book or chapter of book close
• CH close
• Archive ouverte UNIGE close
• Hyper Article en Ligne close
• Digital Humanities and Cultural Heritage close

Almost entirely overlooked throughout the 20th century, neglected by contemporary medical manuals, the clitoris has gradually returned centre stage thanks to Western feminism.

We present the HIPE-2022 shared task on named entity processing in multilingual historical documents. Following the success of the first CLEF-HIPE-2020 evaluation lab, this edition confronts systems with the challenges of dealing with more languages, learning domain-specific entities, and adapting to diverse annotation tag sets. HIPE-2022 is part of the ongoing efforts of the natural language processing and digital humanities communities to adapt and develop appropriate technologies to efficiently retrieve and explore information from historical texts. On such material, however, named entity processing techniques face the challenges of domain heterogeneity, input noisiness, dynamics of language, and lack of resources. In this context, the main objective of the evaluation lab is to gain new insights into the transferability of named entity processing approaches across languages, time periods, document types, and annotation tag sets.

We consider methods for handling incomplete (elliptical) utterances in spoken phraselators, and describe how they have been implemented inside BabelDr, a substantial spoken medical phraselator. The challenge is to extend the phrase matching process so that it is sensitive to preceding dialogue context. We contrast two methods, one using limited-vocabulary strict grammar-based speech and language processing and one using large-vocabulary speech recognition with fuzzy grammar-based processing, and present an initial evaluation on a spoken corpus of 821 context-sentence/elliptical-phrase pairs. The large-vocabulary/fuzzy method strongly outperforms the limited-vocabulary/strict method over the whole corpus, though it is slightly inferior for the subset that is within grammar coverage. We investigate possibilities for combining the two processing paths, using several machine learning frameworks, and demonstrate that hybrid methods strongly outperform the large-vocabulary/fuzzy method.

With the growing number of electronic health record data, clinical NLP tasks have become increasingly relevant to unlock valuable information from unstructured clinical text. Although the performance of downstream NLP tasks, such as named-entity recognition (NER), in English corpus has recently improved by contextualised language models, less research is available for clinical texts in low resource languages. Our goal is to assess a deep contextual embedding model for Portuguese, so called BioBERTpt, to support clinical and biomedical NER. We transfer learned information encoded in a multilingual-BERT model to a corpora of clinical narratives and biomedical-scientific papers in Brazilian Portuguese. To evaluate the performance of BioBERTpt, we ran NER experiments on two annotated corpora containing clinical narratives and compared the results with existing BERT models. Our in-domain model outperformed the baseline model in F1-score by 2.72%, achieving higher performance in 11 out of 13 assessed entities. We demonstrate that enriching contextual embedding models with domain literature can play an important role in improving performance for specific NLP tasks. The transfer learning process enhanced the Portuguese biomedical NER model by reducing the necessity of labeled data and the demand for retraining a whole new model.

Web accessibility has only recently begun to be considered as a key component in the task of the web localiser and, crucially, in the assessment of localisation quality. The ALMA research project (Approaching Localisation by Means of Accessibility) seeks to address this gap by gradually but comprehensively introducing accessibility awareness, issues and perspectives in the principles and procedures of localisation. One of the approaches of ALMA focuses on localiser education and aims at both integrating web accessibility as content to be transferred in the process of localisation and as a methodological way of rethinking website analysis and interlingual, intercultural, intersemiotic transformation. This would allow localisation students to observe the interrelation between the different semiotic, temporal, spatial or ergodic elements coded in the product, with the aim of being perceived, understood and operated by users through different modalities, senses, capacities and technologies. In this chapter, the specific example of culture and heritage websites is used to illustrate how the social and technological dimensions of multimodal translation, localisation and accessibility converge. By exploring the interrelation of web accessibility, localiser education, Universal Design for Learning, and culture and heritage websites, we conclude that such combination can provide a critical opportunity to enhance accessibility and learning at various levels: as an outcome of localisation training (more accessible multilingual culture and heritage websites), as a motivational driver for all students to access and be engaged in education, as an accessibility-aware mindset and methodology (better and deeper access to training materials), as well as an excellent interdisciplinary tool.

Recurrent neural networks (RNNs) have achieved impressive results in a variety of linguistic processing tasks, suggesting that they can induce non-trivial properties of language. We investigate here to what extent RNNs learn to track abstract hierarchical syntactic structure. We test whether RNNs trained with a generic language modeling objective in four languages (Italian, English, Hebrew, Russian) can predict long-distance number agreement in various constructions. We include in our evaluation nonsensical sentences where RNNs cannot rely on semantic or lexical cues ("The colorless green ideas I ate with the chair sleep furiously"), and, for Italian, we compare model performance to human intuitions. Our language-model-trained RNNs make reliable predictions about long-distance agreement, and do not lag much behind human performance. We thus bring support to the hypothesis that RNNs are not just shallow-pattern extractors, but they also acquire deeper grammatical competence. Accepted to NAACL 2018

In this paper, we investigate the task of learning word embeddings from very sparse data in an incremental, cognitively-plausible way. We focus on the notion of ‘informativeness’, that is, the idea that some content is more valuable to the learning process than other. We further highlight the challenges of online learning and argue that previous systems fall short of implementing incrementality. Concretely, we incorporate informativeness in a previously proposed model of nonce learning, using it for context selection and learning rate modulation. We test our system on the task of learning new words from definitions, as well as on the task of learning new words from potentially uninformative contexts. We demonstrate that informativeness is crucial to obtaining state-of-the-art performance in a truly incremental setup.

Emotion recognition is usually achieved by collecting features (physiological signals, events, facial expressions, etc.) to predict an emotional ground truth. This ground truth is arguably unreliable due to its subjective nature. In this paper, we introduce a new approach to measure the magnitude of an emotion in the latent space of a Neural Network without the need for a subjective ground truth. Our data consists of physiological measurements during video gameplay, game events, and subjective rankings of game events for the validation of our model. Our model encodes physiological features into a latent variable which is then decoded into video game events. We show that the events are ranked in the latent space similarly to the participants' subjective ranks. For instance, our model's ranking is correlated (Kendall $\tau$ of 0.91) with the predictability rankings.

Au Néolithique, les montagnes sont exploitées pour leurs ressources minérales, cynégétiques et pastorales. À partir de 5 500 ans avant notre ère, les premières communautés agropastorales atteignent les Alpes depuis le nord de l’Italie et la vallée du Rhône et s’établissent dans les massifs subalpins comme dans les Alpes internes. Les études archéobotaniques (analyse des macrorestes végétaux, principalement des graines, des fruits et des charbons de bois) permettent de comprendre l’économie végétale de ces communautés néolithiques : quelles espèces, sauvages ou cultivées, étaient récoltées pour le fourrage, pour construire, se nourrir, se soigner, se chauffer ? Les données de cinq sites néolithiques nous indiquent les différentes façons dont ces populations ont exploité leur territoire en tirant profit des ressources de divers biotopes, de l’étage collinéen à l’étage alpin, contribuant ainsi à mieux comprendre la mobilité verticale au Néolithique en contexte alpin. During the Neolithic, mountains were exploited for their mineral, hunting and pastoral resources. The first agro-pastoral communities reached the Alps from Northern Italy and the Rhone valley and settled in the subalpine massifs and in the internal Alps. Archeobotanical studies (plant macroremains and charcoal analysis) conducted at five sites allow us to understand the plant economy of these Neolithic communities: they determine which crops were cultivated, used as fodder, or gathered for consumption, medicine or other purpose, such as firewood. In the present paper, we support that the use of plant resources and the exploitation of territory are very different for the same period from one region to another, depending on the activities carried out at each site but also on cultural backgrounds. Archeobotanical data indicate how these people took resources from various plant associations growing from the colline to the subalpine level, and thus contribute to the understanding of vertical mobility in alpine contexts.