- Corpora - ELRA lists

July 2023 Newsletter - LDC
by Penn LDC 17 Jul '23

17 Jul '23

In this newsletter: LanguageArc featured in Babel magazine Fall 2023 LDC Data Scholarship Program New publications: Mixer 7 Spanish Speech<https://catalog.ldc.upenn.edu/LDC2023S04> LORELEI Thai Representative Language Pack<https://catalog.ldc.upenn.edu/LDC2023T08> ________________________________ LanguageArc featured in Babel magazine The May 2023 edition of Babel<https://cloud.3dissue.com/18743/41457/106040/BabelNo43/index.html> (The Language Magazine) features an article about LDC's citizen science portal LanguageArc<https://languagearc.com/> (Language Analysis Research Community) and the diverse projects available there that utilize a variety of novel incentives to supplement traditional methods of creating data resources. Consider LanguageArc for your next collection project. Note: a subscription is necessary to view the article. Fall 2023 LDC Data Scholarship Program Student applications for the Fall 2023 LDC Data Scholarship program are being accepted now through September 15, 2023. This program provides eligible students with no-cost access to LDC data. Students must complete an application consisting of a data use proposal and letter of support from their advisor. For application requirements and program rules, visit the LDC Data Scholarships page<https://www.ldc.upenn.edu/language-resources/data/data-scholarships>. ________________________________ New publications: Mixer 7 Spanish Speech<https://catalog.ldc.upenn.edu/LDC2023S04> was developed by LDC and contains 9,600 hours of audio recordings of interviews, transcript readings, and conversational telephone speech involving 191 distinct native Spanish speakers. This material was collected by LDC in 2011-2012 as part of the Mixer project, and the recordings were used in the 2012 NIST SRE test set. Recruited speakers were connected through a robot operator to carry on casual conversations on a pre-set topic lasting up to 10 minutes. Participants also visited LDC's human subjects collection lab equipped with a 14-microphone array where they participated in interviews and transcript readings and conducted up to 3 telephone calls under varying conditions. Selected speaker metadata was also collected. 2023 members can access this corpus through their LDC accounts. This corpus is a members-only release and is not available for non-member licensing. Contact ldc(a)ldc.upenn.edu<mailto:ldc@ldc.upenn.edu> for information about membership. * LORELEI Thai Representative Language Pack<https://catalog.ldc.upenn.edu/LDC2023T08> is comprised of over 39 million words of Thai monolingual text, 2.85 million words of found Thai-English parallel text, and 141,000 Thai words translated from English data. Over 186,000 words were annotated for named entities and more than 25,000 words were annotated for entity discovery and linking and situation frames (identifying entities, needs, and issues). Data was collected from discussion forum, news, reference, social network, and weblogs. The LORELEI (Low Resource Languages for Emergent Incidents) program was concerned with building human language technology for low resource languages in the context of emergent situations. Representative languages were selected to provide broad typological coverage. The knowledge base for entity linking annotation is available separately as LORELEI Entity Detection and Linking Knowledge Base (LDC2020T10)<https://catalog.ldc.upenn.edu/LDC2020T10>. 2023 members can access this corpus through their LDC accounts. Non-members may license this data for a fee. To unsubscribe from this newsletter, log in to your LDC account<https://catalog.ldc.upenn.edu/login> and uncheck the box next to "Receive Newsletter" under Account Options or contact LDC for assistance.

1 0

DHd2024 Final Call for Papers
by Haider, Thomas 17 Jul '23

17 Jul '23

Dear colleagues, This is the final call for papers for DHd2024 in Passau. The deadline is already in a few days on 19.07.2023. Important dates: 19.07.2023 Submission Deadline 02.11.2023 Notification of acceptance 04.12.2023 Submission of camera ready version 04.12.2023 Start of registration 15.02.2024 End of registration 26.02.2024 - 01.03.2024 Conference in Passau Website: https://dhd2024.dig-hum.de/ Call for Papers: https://dhd2024.dig-hum.de/call-for-papers/ (also in English) Conftool: https://www.conftool.net/dhd2024/ Links for submission (Convalidator): https://dhd2024.dig-hum.de/submission/ Paper Guidelines: https://dhd2024.dig-hum.de/guidelines/ If you already had a Conftool account for DHd2023 in Trier, your registration details are the same. Please check and update your details. Best regards Thomas Haider ---- Dr. phil. Thomas Nikolaus Haider Digital Humanities and Multilingual Computational Linguistics University of Passau

1 0

Second Call for papers: The First Bangla Language Processing Workshop (BLP 2023) Co-located with EMNLP 2023 in Singapore
by Firoj Alam 17 Jul '23

17 Jul '23

***Apologies for Cross-Posting*** Second Call for Papers: The first BNLP workshop aims to provide a forum for the NLP, speech and multimodal communities to share and discuss their ongoing work with the international community. We particularly focus on Bangla, which is a low-resource language, and assess its current state-of-the-art and discuss strategies to make further progress in NLP, speech and multimodal research. Through this workshop, we plan to bring researchers together to come up with frameworks and strategies that can later support other low-resource languages. We encourage researchers to submit their papers focusing on novel methodologies and resources that help towards the progress of Bangla and other low resource languages. Novel methodologies include, but are not limited to, zero-shot learning, unsupervised learning, and simple yet effective methods applicable to low-computation scenarios. We invite original research papers from a wide range of topics, including but not limited to: Natural Language Processing: Corpus and Resource Development, Language Modeling, Stemmer, POS Tagger, Named Entity Recognition, Relation Extraction, Spell and Grammar Checker, Question Answering, Semantics, Text Summarization, Machine Translation, Sentiment Analysis. Speech Processing: Speech Synthesis and Spoken Language Generation, Speech Recognition, Phonetics, Phonology, and Prosody, Spoken Dialog and Conversational System, Speaker and Language Detection. Multimodality: OCR - Handwriting, Printed Document, Sign Language Detection. Human Computer Interaction: Software for Disabled People, Multimodal HCI for Bangla. Important dates: Workshop paper due: 1 September 2023 Notification of acceptance: 6 October 2023 Camera-ready papers due: 18 October 2023 Workshop dates: 6-7 December 2023 All deadlines are 11:59pm anywhere on Earth (AoE). Submission Details: Papers must describe original, completed or in-progress, and unpublished work. All papers will be refereed through a double-blind peer review process by multiple reviewers with final acceptance decisions made by the workshop organizers. Accepted papers will be given up to 9 pages (for full papers), 5 pages (for short papers and posters) in the workshop proceedings, and will be presented as oral paper or poster. We are seeking submissions under the following category - Full papers (8 pages) - Short papers (work in progress, innovative ideas/proposals: 4 pages) - Shared task paper (4 pages) Both long and short papers must follow the EMNLP 2023 two-column format, using the supplied official templates [1]. The templates can be downloaded in style files and formatting. Please do not modify these style files, nor should you use templates designed for other conferences. Submissions that do not conform to the required styles, including paper size, margin width, and font size restrictions, will be rejected without review. Verification to guarantee conformance to publication standards, we will be using the ACL pubcheck tool [2]. The PDFs of camera-ready papers must be run through this tool prior to their final submission, and we recommend its use also at submission time. Submissions are open to all, and are to be submitted anonymously. For the anonymity, double-blind submission and reproducibility criteria please follow the EMNLP 2023 instructions [3]. If you have published in the field previously, and are interested in helping out in the program committee to review papers, please fill up this form <https://forms.gle/1WUYQjWT9UuqioX48> [4]. Submission portal: TBA Workshop Organizers: Firoj Alam, Qatar Computing Research Institute, HBKU, Qatar Sudipta Kar, Amazon Alexa AI, USA Shammur Absar Chowdhury, Qatar Computing Research Institute, HBKU, Qatar Farig Sadeque, BRAC University, Bangladesh Ruhul Amin, Fordham University, USA [1] https://2023.emnlp.org/calls/style-and-formatting/ [2] https://github.com/acl-org/aclpubcheck [3] https://2023.emnlp.org/calls/main_conference_papers/ [4] https://forms.gle/1WUYQjWT9UuqioX48 The Organizers

1 0

Six PhD students and one postdoc: Neurosymbolic Models of Language, Vision, and Action - Saarbrücken, Germany
by Alexander Koller 16 Jul '23

16 Jul '23

The Research Training Group 2853 “Neuroexplicit Models of Language, Vision, and Action” is looking for 6 PhD students and 1 postdoc October 2023 or later Neuroexplicit models combine neural and human-interpretable (“explicit”) models in order to overcome the limitations that each model class has separately. They include neurosymbolic models, which combine neural and symbolic models, but also e.g. combinations of neural and physics-based models. In the RTG, we will improve the state of the art in natural language processing (“Language”), computer vision (“Vision”), and planning and reinforcement learning (“Action”) through the use of neuroexplicit models and investigate the cross-cutting design principles of effective neuroexplicit models (“Foundations”). The RTG is scheduled to grow to a total of 24 PhD students and one postdoc by 2025. Through the inclusion of ~20 further PhD students and postdocs funded from other sources, it will be one of the largest research centers on neuroexplicit or neurosymbolic models in the world. The RTG brings together researchers at Saarland University, the Max Planck Institute for Informatics, the Max Planck Institute for Software Systems, the CISPA Helmholtz Center for Information Security, and the German Research Center for Artificial Intelligence (DFKI). All of these institutions are colocated on the same campus in Saarbrücken, Germany. The positions are funded as follows: • PhD students will be funded for up to four years at the TV-L E13 100% pay scale. You should have or be about to complete an MSc degree in computer science or a related field and have demonstrated expertise in one of the research areas of the RTG, e.g. through an excellent Master’s thesis or relevant publications. • The postdoc will initially be funded for three years, with the possibility of extension up to five years, at the TV-L E13 100% pay scale. As the RTG postdoc, you will pursue your own research agenda in the field of neuroexplicit models and work with the PhD students to identify and pursue opportunities for collaborative research. You should have or be about to complete a PhD in computer science or a related field and have demonstrated your expertise in one or more of the RTG’s research areas through publications in top venues. The RTG is part of the Saarland Informatics Campus, one of the leading centers for research in computer science, artificial intelligence, and natural language processing in Europe. The Saarland Informatics Campus brings together 900 researchers and 2500 students from 81 countries. The CISPA Helmholtz Center, located on the same campus, is home to an additional 350 researchers and on track to grow to 800 by 2026. Researchers at SIC and CISPA are part of the ELLIS network and have been awarded more than 35 ERC grants. Each PhD student in the RTG will be jointly supervised by two PhD advisors from the list of Principal Investigators below. Each student will freely define their own research topic; we encourage the choice of topics that cross the traditional boundaries of research fields. Students may be affiliated with Saarland University or with one of the participating institutes. Vera Demberg, Saarland University - Computational Linguistics Jörg Hoffmann, Saarland University - AI Planning Eddy Ilg, Saarland University - Computer Vision, Machine Learning Dietrich Klakow, Saarland University - Natural Language Processing Alexander Koller, Saarland University - Computational Linguistics Bernt Schiele, MPI for Informatics - Computer Vision, Machine Learning Philipp Slusallek, DFKI and Saarland University - Computer Graphics, Artificial Intelligence Christian Theobalt, MPI for Informatics - Visual Computing, Machine Learning Mariya Toneva, MPI for Software Systems - Computational Neuroscience, Machine Learning Isabel Valera, Saarland University - Machine Learning Jilles Vreeken, CISPA - Machine Learning, Causality Joachim Weickert, Saarland University - Mathematical Data Analysis Verena Wolf, DFKI and Saarland University - Modeling and Simulation, Reinforcement Learning Ellie Pavlick, Brown University and Google AI, will join us regularly as a Mercator Fellow. Please send your application by 31 May 2023 to bewerbung(a)uni-saarland.de. Include the reference number W2298 for the postdoc position and the reference number W2299 for the PhD positions. We aim to conduct job interviews in July (for a start in October) and September (for a later start). The legally binding version of this job ad is at https://www.uni-saarland.de/fileadmin/upload/verwaltung/stellen/Wissenschaf… (postdoc) and https://www.uni-saarland.de/fileadmin/upload/verwaltung/stellen/Wissenschaf… (PhD), respectively. For details on what materials to submit with your application and all other information about the RTG, please see our website: https://www.neuroexplicit.org/jobs/#phd-2023

2 1

[Deadline extension] PHRAME closing event - Perugia, 27/28.11.2023
by Luciana Forti 15 Jul '23

15 Jul '23

PHRAME: Phraseological complexity measures in learner Italian. Integrating eye tracking, computational and learner corpus methods to develop second language pedagogical resources Closing event Perugia, 27-28 November 2023 SECOND CALL FOR PAPERS We invite contributions including, but not limited to, the following areas of research: - Corpus-based approaches in the study of L2 phraseology - Psycholinguistic approaches in the study of L2 phraseology - Computational and NLP approaches in the study of L2 phraseology - Complexity theory and second language acquisition studies - Mixed-methods approaches in the study of L2 phraseology - Syllabus design for the development of phraseological competence - Testing and assessing L2 phraseological competence Keynote speakers of the event will be: - Philip Durrant (University of Exeter) - Gabriele Pallotti (University of Modena and Reggio Emilia) - Anna Siyanova-Chanturia (University of Wellington and Ocean University of China) Guidelines - The event will be in-person only, with live-streaming. - Paper proposals may refer to languages other than Italian. - The abstracts for full paper presentations should be written in English. They should not exceed 500 words, references included. The abstract should be sent to progettophrame(a)gmail.com in a Word document containing the following information: - Title of the proposed presentation - Name of the author(s) - Affiliation of the author(s) - The presented papers will be collected and published. - Registration for in-person participation will open on 4th September and close on 10th November 2023. - The official languages of the event are Italian and English. Important dates 16 July 2023 31 July 2023: Abstract submission closes. 30 August 2023: Abstract reviews assigned. 17 September 2023: Abstract reviews completed. 22 September 2023: Notification to authors. 10 November 2023: Registration for in-person participation closes. 27-28 November 2023: PHRAME closing event. Any questions regarding the call for papers and the event may be directed to progettophrame(a)gmail.com. Scientific and organising Committee: Giulio Biondi, Veronica D’Alesio, Irene Fioravanti, Luciana Forti, Valentina Franzoni, Sabine Koesters Gensini, Francesca La Russa, Francesca Malagnini, Alfredo Milani, Maria Roccaforte, Valentino Santucci, Stefania Spina, Fabio Zanda. References - Altenberg, B. (1998). On the phraseology of spoken English: The evidence of recurrent word-combinations. In A. P. Cowie (Ed.), Phraseology: Theory, analysis and applications (pp. 101–122). Oxford University Press. - Paquot, M. (2018). Phraseological Competence: A Missing Component in University Entrance Language Tests? Insights From a Study of EFL Learners’ Use of Statistical Collocations. Language Assessment Quarterly, 1–15. - Sinclair, J. M. (1991). Corpus, Concordance, Collocation. Oxford University Press. - Siyanova-Chanturia, A., & Pellicer-Sánchez, A. (2019). Understanding Formulaic Language A Second Language Acquisition Perspective. Routledge. Web page of the event: PHRAME: Phraseological complexity Measures in learner Italian (Closing event) | | | | | | | | | | | PHRAME: Phraseological complexity Measures in learner Italian (Closing e... Second Call for Papers We invite contributions including, but not limited to, the following areas of research: C... | | |

1 0

Call for Participation: Challenge on Medical Video Question Answering at TRECVID 2023
by Deepak Gupta 15 Jul '23

15 Jul '23

Dear colleagues and friends, This year, we are organizing the MedVidQA <https://medvidqa.github.io/>challenge with TRECVID 2023 <https://www-nlpir.nist.gov/projects/tv2023/index.html>. This challenge aims at developing models for (1) retrieving the relevant videos and locating the visual answer in those videos for the medical or health-related question and (2) generating the medical instructional questions from the video segments. Following the success of the 1st MedVidQA shared task <https://aclanthology.org/2022.bionlp-1.25/>, MedVidQA at TRECVID 2023 expanded the tasks and introduced a new track considering language-video understanding and generation. This track is comprised of two main tasks Video Corpus Visual Answer Localization (VCVAL) and Medical Instructional Question Generation (MIQG). For more details, please visit the challenge website ( https://medvidqa.github.io/) and TRECVID 2023 website ( https://www-nlpir.nist.gov/projects/tv2023/index.html). The link for submission: - Task 1 (VCVAL): https://codalab.lisn.upsaclay.fr/competitions/13445 <https://codalab.lisn.upsaclay.fr/competitions/13546> - Task 2 (MIQG): https://codalab.lisn.upsaclay.fr/competitions/13546 *Important Dates* - *Release of the training and validation datasets:* April 30, 2023 - *Release of the video corpus:* May 12, 2023 - *Release of the test sets:* July 14, 2023 - *Run submission deadline:* August 4, 2023 - *Release of the official results:* September 29, 2023 We look forward to your participation in MedVidQA at TRECVID 2023. Join our Google Group <https://groups.google.com/g/trecvid-medvidqa2023> for important updates! If you have any questions, ask in our Google Group <https://groups.google.com/g/trecvid-medvidqa2023> or email <deepak.gupta(a)nih.gov> us. Thank you, MedVidQA 2023 Organizers

1 2

Linguamática V15N1 is available!
by Hugo Gonçalo Oliveira 14 Jul '23

14 Jul '23

---------------------------------------------------------------------- Linguamática Volume 15, Número 1 1647-0818 --- http://www.linguamatica.com ---------------------------------------------------------------------- ** Artigos relacionados com a DIP: Desafio de Identificação de Personagens ** Diana Santos, Cristina Mota, Emanoel Pires, Marcia Langfeldt, Rebeca Schumacher Fuão, Roberto Willrich. DIP - Desafio de Identificação de Personagens: objectivo, organização, recursos e resultados https://linguamatica.com/index.php/linguamatica/article/view/399/492 Eckhard Bick. Extração de Informação sobre Personagens Literários em Português https://linguamatica.com/index.php/linguamatica/article/view/397/490 Diana Santos, Cristina Mota. Pais, filhos e outras relações familiares no DIP https://linguamatica.com/index.php/linguamatica/article/view/402/493 Emanoel Pires, Marcia Caetano Langfeldt, Rebeca Schumacher Fuão. Desafios e vantagens do processo de identificação automática do gênero e das profissões das personagens no DIP https://linguamatica.com/index.php/linguamatica/article/view/401/491 Roberto Willrich, Diana Santos. Avaliação no Desafio de Identificação de Personagens https://linguamatica.com/index.php/linguamatica/article/view/398/494 ** Artigos de Investigação / Artículos de Investigación / Research Articles ** David Soares Batista. Extracção de Relações de Apoio e Oposição em Títulos de Notícias de Política em Português https://linguamatica.com/index.php/linguamatica/article/view/386/495 Cássio Faria da Silva, Vânia Paula de Almeida Neris, Helena de Medeiros Caseli. Classificação da qualidade da argumentação em tweets no domínio da política brasileira https://linguamatica.com/index.php/linguamatica/article/view/387/496 ** Novas Perspectivas / Nuevas Perspectivas / New Perspectives ** Átila Augusto Soares Vital. A compilação e a análise de métricas textuais de um corpus de redações https://linguamatica.com/index.php/linguamatica/article/view/393/497

1 0

18 month postdoc position in Natural Language Processing at the University of Copenhagen
by Desmond Elliott 14 Jul '23

14 Jul '23

The Natural Language Processing Section at the Department of Computer Science at University of Copenhagen is advertising an 18 month position for a Postdoctoral Researcher in Natural Language Processing. The position is funded by a VILLUM Young Investigator Grant held by the principal investigator, Desmond Elliott. The overall goal of the project is to develop a new family of language models that can process any written language by rendering text as images, which allows the models to learn from the visual similarities between written languages, facilitating effective transfer to lower-resource or unseen languages. The Natural Language Processing Section provides a strong, international and diverse environment for research within core as well as emerging topics in natural language processing, natural language understanding, computational linguistics and multi-modal language processing. It is housed within the main Science Campus, which is centrally located in Copenhagen. Further information about research at the Department is available here: https://di.ku.dk/english/research/. The application deadline is 31 August 2023, with a preferred start date of January 2024, or as soon as possible thereafter. More information and your application can be submitted here: https://di.ku.dk/english/about/vacancies/postdoc-in-natural-language-proces… Informal enquiries about the positions can be made to Desmond Elliott, Department of Computer Science, University of Copenhagen, e-mail: de(a)di.ku.dk.

1 0

Three PhD students, one postdoc on neurosymbolic models
by Alexander Koller 14 Jul '23

14 Jul '23

The Research Training Group 2853 “Neuroexplicit Models of Language, Vision, and Action” is looking for 3 PhD students and 1 postdoc November 2023 or later Neuroexplicit models combine neural and human-interpretable (“explicit”) models in order to overcome the limitations that each model class has separately. They include neurosymbolic models, which combine neural and symbolic models, but also e.g. combinations of neural and physics-based models. In the RTG, we will improve the state of the art in natural language processing (“Language”), computer vision (“Vision”), and planning and reinforcement learning (“Action”) through the use of neuroexplicit models and investigate the cross-cutting design principles of effective neuroexplicit models (“Foundations”). The RTG is scheduled to grow to a total of 24 PhD students and one postdoc by 2025. Through the inclusion of ~20 further PhD students and postdocs funded from other sources, it will be one of the largest research centers on neuroexplicit or neurosymbolic models in the world. The RTG brings together researchers at Saarland University, the Max Planck Institute for Informatics, the Max Planck Institute for Software Systems, the CISPA Helmholtz Center for Information Security, and the German Research Center for Artificial Intelligence (DFKI). All of these institutions are colocated on the same campus in Saarbrücken, Germany. In a previous round of recruiting, we have already filled three PhD positions, leaving three positions free for applicants with a later MSc graduation date. The positions are funded as follows: • PhD students will be funded for up to four years at the TV-L E13 100% pay scale. You should have or be about to complete an MSc degree in computer science or a related field and have demonstrated expertise in one of the research areas of the RTG, e.g. through an excellent Master’s thesis or relevant publications. • The postdoc will initially be funded for three years, with the possibility of extension up to five years, at the TV-L E13 100% pay scale. As the RTG postdoc, you will pursue your own research agenda in the field of neuroexplicit models and work with the PhD students to identify and pursue opportunities for collaborative research. You should have or be about to complete a PhD in computer science or a related field and have demonstrated your expertise in one or more of the RTG’s research areas through publications in top venues. The RTG is part of the Saarland Informatics Campus, one of the leading centers for research in computer science, artificial intelligence, and natural language processing in Europe. The Saarland Informatics Campus brings together 900 researchers and 2500 students from 81 countries. The CISPA Helmholtz Center, located on the same campus, is home to an additional 350 researchers and on track to grow to 800 by 2026. Researchers at SIC and CISPA are part of the ELLIS network and have been awarded more than 35 ERC grants. Each PhD student in the RTG will be jointly supervised by two PhD advisors from the list of Principal Investigators below. Each student will freely define their own research topic; we encourage the choice of topics that cross the traditional boundaries of research fields. Students may be affiliated with Saarland University or with one of the participating institutes. Vera Demberg, Saarland University - Computational Linguistics Jörg Hoffmann, Saarland University - AI Planning Eddy Ilg, Saarland University - Computer Vision, Machine Learning Dietrich Klakow, Saarland University - Natural Language Processing Alexander Koller, Saarland University - Computational Linguistics Bernt Schiele, MPI for Informatics - Computer Vision, Machine Learning Philipp Slusallek, DFKI and Saarland University - Computer Graphics, Artificial Intelligence Christian Theobalt, MPI for Informatics - Visual Computing, Machine Learning Mariya Toneva, MPI for Software Systems - Computational Neuroscience, Machine Learning Isabel Valera, Saarland University - Machine Learning Jilles Vreeken, CISPA - Machine Learning, Causality Joachim Weickert, Saarland University - Mathematical Data Analysis Verena Wolf, DFKI and Saarland University - Modeling and Simulation, Reinforcement Learning Ellie Pavlick, Brown University and Google AI, will join us regularly as a Mercator Fellow. Please send your application by 31 July 2023 to apply(a)neuroexplicit.org. Include the reference number W2350 for the postdoc position and the reference number W2351 for the PhD positions. We aim to conduct job interviews in September, ideally in person in Saarbrücken, Germany. Make sure to check the RTG website for details on the application process and what materials to include in the application: https://www.neuroexplicit.org/jobs For all further information about the RTG, check out our website: https://www.neuroexplicit.org/

1 0

Call for papers - AAC/CFP Corpus 26 - 2025
by Luca Pallanti 13 Jul '23

13 Jul '23

AAC/CFP Corpus 26 - 2025 - https://journals.openedition.org/corpus/ <https://journals.openedition.org/corpus/> Background noise or added value? Managing noise during computer processing of linguistic corpora Elisa Gugliotta, Luca Pallanti, Olivier Kraif, Iris Fabry et Martina Barletta (eds.) -------FRENCH VERSION BELOW----- The increasing influence of NLP-related methodologies on corpus linguistics has compelled researchers to reassess their practices for managing noise and its impact on research results (Fuchs & Habert, 2004; Léon, 2018; Zalmout et al., 2018). Whether working with long-diachronic corpora (e.g., medieval French), dialectal corpora with limited resources (e.g., oral or written texts in dialectal Arabic, cf. Arabizi), or corpora of texts deviating from the norm (e.g., learner corpora), conducting noise analysis becomes an essential step in drawing linguistic conclusions from the available data (Molinelli & Putzu, 2015; Scaglione, 2018; Litosseliti, 2018). This special issue of Corpus builds upon a workshop held in April 2023 (https://je-bruit-corpus.sciencesconf.org/) and offers an opportunity to examine noise management methods in the fields of NLP and corpus linguistics, as well as their impact on the quality of linguistic data (Kraif & Ponton, 2007; Goutte et al., 2012; Zeroual, 2018). The fundamental inquiries in any linguistic study revolve around defining the research object, understanding the nature of the data, and determining ways to preserve its inherent characteristics throughout the various processing steps (such as lemmatisation, normalisation, labelling, etc.) (Sarrica et al., 2016). Hence, selecting appropriate methods for identifying and controlling noise becomes crucial throughout the entire process, from data collection to the archiving phase, and from data preparation to annotation (Egbert & Baker, 2019). The definition of noise itself is diverse and far from self-evident. In the field of NLP alone, this term encompasses a wide range of highly heterogeneous phenomena, including web peritexts - such as hyperlinks, menus and computer codes - as well as code switching and instances of spelling or grammatical errors that punctuate productions (Al Sharou et al., 2021). This special issue aims to delve into the definition of noise, from a linguistic perspective, and the practices employed by researchers to mitigate the biases that can arise from it. These practices are implemented during collection, recording, and annotation of data. The question of noise inevitably emerges at each stage of the empirical process involved in data construction and analysis: 1. Noise during data collection and recording If one accepts the postulate that "linguistic data is a result" (Benveniste, 1966), decoding the noise stemming from data collection and recording becomes crucial. Depending on the research object, various factors may contribute to data alteration, including the researcher's preconceptions or the biases introduced by an OCR system (Jentsch & Porada, 2020). The key challenge lies in predicting or identifying the potential biases induced by these factors during the selection and formatting of data. This enables better control over subsequent research stages and ensures greater accuracy in the analysis process. 2. Data preparation and pre-processing The methods employed to refine raw data and prepare it for advanced manipulation can give rise to a significant source of noise (or, conversely, of silence, if noise elimination filters are applied). This is particularly evident during the data normalization process (Al Sharou et al., 2021). When transcribing data or correcting errors, researchers must make choices that inevitably influence the nature of the data, either by reducing or enriching its content. As a result, it becomes essential to anticipate the consequences of the transformations introduced by data processing methods (Tanguy, 2012). 3. The annotation process and metadata Initially, corpus annotation aims to enrich the data by categorizing units through a labelling process, depending on the developed analysis model (Péry-Woodley et al., 2011). However, while this process has the potential to introduce noise, it can result in detrimental silence (when missing or erroneous labels lead to incomplete results during data analysis or querying). The concept of metadata also raises questions: does categorizing data transform it into something different? Furthermore, does the absence of agreement or low agreement in annotations produced by humans reflect inter-individual variations akin to noise, or does it stem from the inherent vagueness of the categorizations themselves? *** At each and every step of the process, key methodological questions arise: what threshold can be considered acceptable for noise? How can we differentiate between noise and methodological bias? Is it possible to estimate noise without a ground truth? Which statistical tools are specific to corpus studies and enable the definition of confidence intervals? How can we strike a balance to prevent the noise resulting from compromising research outcomes? *** Proposals for articles may address these topics from a general point of view, offering a theoretical and methodological perspective. Alternatively, they can be based on one or more case studies that focus on specific observations, while highlighting the noise management methods employed throughout the study. References Al Sharou, K., Li, Z., & Specia, L. (2021). Towards a Better Understanding of Noise in Natural Language Processing. Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021), 5362. https://aclanthology.org/2021.ranlp-1.7 Benveniste, É. (1966). Problèmes de linguistique générale. Gallimard. Egbert, J., & Baker, P. (Eds.). (2019). Using corpus methods to triangulate linguistic analysis. Routledge. Fuchs, C., & Habert, B. (2004). Le traitement automatique des langues : Des modèles aux ressources. Le Français Moderne - Revue de linguistique Française, CILF (conseil international de la langue française), LXXII: 1, online. Goutte, C., Carpuat, M., & Foster, G. (2012). The impact of sentence alignment errors on phrase-based machine translation performance. In Proceedings of the 10th Conference of the Association for Machine Translation in the Americas: Research Papers. Jentsch, P., & Porada, S. (2020). From Text to Data : Digitization, Text Analysis and Corpus Linguistics. In S. Schwandt (Éd.), Digital Humanities Research (1re éd., Vol. 1, p. 89128). transcript Verlag / Bielefeld University Press. https://doi.org/10.14361/9783839454190-004 Kraif, O., & Ponton, C. (2007). Du bruit, du silence et des ambiguïtés : Que faire du TAL pour l'apprentissage des langues ? TALN 2007, 143152. https://hal.archives-ouvertes.fr/hal-01073706 Léon, J. (2018). Tal et linguistique : Application, expérimentation, instrumentalisation. ELA. Etudes de linguistique appliquee, 2(190), 195203. Litosseliti, L. (Ed.). (2018). Research methods in linguistics. Bloomsbury Publishing. Molinelli, P., & Putzu, I. (2015). Modelli epistemologici, metodologie della ricerca e qualità del dato. Dalla linguistica storica alla sociolinguistica storica. Franco Angeli. Péry-Woodley, M.-P., Afantenos, S. D., Ho-Dac, L.-M., & Asher, N. (2011). La ressource ANNODIS, un corpus enrichi d'annotations discursives. TAL, 52(3), 71101. Sarrica, M., Mingo, I., Mazzara, B., & Leone, G. (2016). The effects of lemmatization on textual analysis conducted with IRaMuTeQ: results in comparison. JADT2016: 13ème Journées Internacionales d'Analyse Statistique de Données Textuelles. Scaglione, F. (2018). "Lavorare"; il dato linguistico: Prospettive e limiti. Alcune considerazioni dall'esperienza dell'Atlante Linguistico della Sicilia (ALS). In G. Sampino (Éd.), Atti del convegno internazionale dei dottorandi (p. 101122). Tanguy, L. (2012). Complexification des données et des techniques en linguistique : contribution du TAL aux solutions et aux problèmes. HDR dissertation, Université de Toulouse 2 - le Mirail. Zalmout, N., Erdmann, A., & Habash, N. (2018). Noise-robust morphological disambiguation for dialectal Arabic. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) (pp. 953-964). Zeroual, I. (2018). Building Arabic Corpora: Concepts, Methodologies, Tools, and Experiments (Doctoral dissertation, University of Maryland, USA). Retro-planning * July 2023: call for publications. * November 2023: pre-selection based on article summaries. * March 2024: article submission deadline. * June 2024: response to the authors. * June-October 2024: review process with authors to submit the final version of the article. * November-December 2024: editing process. * January 2025: publication. Please note that this retro-planning outlines a general timeline and may vary depending on the specific publication requirements. Abstract submission * Your abstract should be no longer than 1,500 words, including bibliographical references. * Please submit your abstracts by November 6, 2023 to elisa.gugliotta(a)ilc.cnr.it and luca.pallanti(a)univ-lyon2.fr. ----- FRENCH VERSION------ Bruit de fond ou valeur ajoutée ? Gérer le bruit lors des traitements informatiques des corpus linguistiques Sous la direction de Elisa Gugliotta, Luca Pallanti, Olivier Kraif, Iris Fabry et Martina Barletta English version below L'influence croissante des méthodologies liées au TAL sur la linguistique de corpus oblige les chercheurs à réinterroger les pratiques de gestion du bruit et son impact dans les résultats de recherche (Fuchs & Habert, 2004 ; Léon, 2018 ; Zalmout et al., 2018). Qu'il s'agisse de corpus en diachronie longue (ex. français médiéval), de corpus dialectaux aux ressources limitées (ex. textes oraux ou écrits en arabe dialectal, cf. arabizi), ou encore de corpus de textes éloignés de la norme (ex. corpus d'apprenants), l'analyse du bruit est une étape nécessaire pour tirer des conclusions linguistiques des données ainsi évaluées (Molinelli & Putzu, 2015 ; Scaglione, 2018 ; Litosseliti, 2018). Ce numéro thématique de la revue Corpus, qui fait suite à une journée d'étude sur le même thème organisée en avril 2023 (https://je-bruit-corpus.sciencesconf.org/), sera l'occasion de réfléchir sur les méthodes de gestion du bruit dans les domaines du TAL et de la linguistique de corpus outillée, et à son impact sur la qualité des données linguistiques (Kraif et Ponton, 2007 ; Goutte et al., 2012 ; Zeroual, 2018). Les questions sous-jacentes à toute étude linguistique concernent la définition de l'objet de recherche, la nature des données elles-mêmes, et la manière de préserver autant que possible leurs caractéristiques dans les différents traitements (lemmatisation, normalisation, étiquetage, etc.) (Sarrica et al., 2016). Ainsi, le choix des méthodes d'identification et de contrôle du bruit, de la phase de collecte à celle d'archivage, de la préparation des données à l'annotation, joue un rôle fondamental (Egbert & Baker, 2019). La définition même du bruit est multiple, et ne va pas de soi : dans le seul champ du TAL, ce terme, souvent peu interrogé, désigne des phénomènes variables et très hétérogènes, allant des péritextes du Web - hyperliens, menus et codes informatiques - au code switching, en passant par les erreurs d'orthographe ou de grammaire qui émaillent les productions (Al Sharou et al., 2021). Ce numéro thématique propose de mener une réflexion sur la définition du bruit, dans une perspective linguistique, et sur les pratiques des chercheurs visant à réduire la portée des biais qui en découlent, que ce soit durant la collecte, l'enregistrement ou l'annotation des données. Dans le concret de la recherche, la question du bruit se pose à chaque étape de la démarche empirique de construction et d'analyse des données : 1. Le bruit pendant la collecte et l'enregistrement des données Si l'on accepte le postulat selon lequel " la donnée linguistique est un résultat " (Benveniste, 1966), comment décoder le bruit causé par le recueil des données et leur enregistrement ? En effet, en fonction des objets de recherche, il existe des facteurs potentiels d'altération des données, comme par exemple les préconceptions du chercheur, ou les biais introduits par un système OCR donné (Jentsch & Porada, 2020). L'enjeu consiste alors à prédire ou à déterminer les biais potentiels induits par ces facteurs lors de la sélection et la mise en forme des données pour mieux contrôler les phases de recherche successives. 2. La préparation et le prétraitement des données. Les méthodes choisies pour affiner les données brutes et les rendre disponibles pour des manipulations avancées peuvent représenter une importante source de bruit (ou, au contraire, de silence si on applique un filtre pour éliminer le bruit) : c'est notamment le cas du processus de normalisation des données (Al Sharou et al., 2021). Qu'il s'agisse de transcrire des données ou de corriger des erreurs, le chercheur fait des choix qui impactent nécessairement la nature des données, soit en les réduisant, soit en les enrichissant. Il s'agit donc d'anticiper les conséquences des transformations produites par les méthodes de traitement des données (Tanguy, 2012). 3. Le processus d'annotation et les métadonnées À la base, l'annotation des corpus est une étape visant l'enrichissement des données : en fonction du modèle d'analyse mis au point, le chercheur tente de catégoriser des unités à travers un processus d'étiquetage (Péry-Woodley et al., 2011). Cependant, si d'un côté ce processus peut générer du bruit, de l'autre, il peut être une cause de silence fort préjudiciable aux résultats des recherches et à leur interprétation (des étiquettes absentes ou erronées pouvant générer des résultats lacunaires lors de l'analyse ou du requêtage des données). La notion de métadonnée peut également être mise en cause : catégoriser une donnée signifie-t-il la transformer en quelque chose d'autre ? Par ailleurs, l'absence d'accord ou un faible accord dans les annotations produites par l'humain manifeste-t-il des variations interindividuelles assimilables à du bruit, ou au caractère trop vague des catégorisations en jeu ? *** A chaque étape se posent des questions méthodologiques centrales : à partir de quel seuil peut-on considérer le bruit comme acceptable ? Comment différencier bruit et biais méthodologique ? Comment estimer le bruit sans vérité de terrain ? Quels outils statistiques spécifiques à l'étude des corpus permettent de délimiter des intervalles de confiance ? Comment atteindre l'équilibre nécessaire pour que le bruit causé par les traitements des données ne compromette pas les résultats des recherches ? *** Les propositions d'article pourront aborder ces questions d'un point de vue général, sous un angle théorique et méthodologique, ou s'appuyer sur une ou plusieurs études de cas portant sur des observations particulières, en prenant soin de mettre en lumière les méthodes de gestion du bruit tout au long de l'étude. Retro-planning * Juillet 2023 : publication du l'Appel * Novembre 2023 : pré-sélection sur résumé * Mars 2024 : remise des articles. Juin 2024 : réponse aux auteurs * Juin-octobre 2024 : navette avec les auteurs pour remise de l'article en forme définitive. * Novembre-décembre 2024 : édition. * Janvier 2025 : publication. Soumission des résumés * Votre résumé comptera 1.500 mots au maximum, références bibliographiques inclues. * Merci de soumettre vos résumés pour le 6 novembre 2023 aux adresses elisa.gugliotta(a)ilc.cnr.it et luca.pallanti(a)univ-lyon2.fr Références Al Sharou, K., Li, Z., & Specia, L. (2021). Towards a Better Understanding of Noise in Natural Language Processing. Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021), 5362. https://aclanthology.org/2021.ranlp-1.7 Benveniste, É. (1966). Problèmes de linguistique générale. Gallimard. Egbert, J., & Baker, P. (Eds.). (2019). Using corpus methods to triangulate linguistic analysis. Routledge. Fuchs, C., & Habert, B. (2004). Le traitement automatique des langues : Des modèles aux ressources. Le Français Moderne - Revue de linguistique Française, CILF (conseil international de la langue française), LXXII: 1, online. Goutte, C., Carpuat, M., & Foster, G. (2012). The impact of sentence alignment errors on phrase-based machine translation performance. In Proceedings of the 10th Conference of the Association for Machine Translation in the Americas: Research Papers. Jentsch, P., & Porada, S. (2020). From Text to Data : Digitization, Text Analysis and Corpus Linguistics. In S. Schwandt (Éd.), Digital Humanities Research (1re éd., Vol. 1, p. 89128). transcript Verlag / Bielefeld University Press. https://doi.org/10.14361/9783839454190-004 Kraif, O., & Ponton, C. (2007). Du bruit, du silence et des ambiguïtés : Que faire du TAL pour l'apprentissage des langues ? TALN 2007, 143152. https://hal.archives-ouvertes.fr/hal-01073706 Léon, J. (2018). Tal et linguistique : Application, expérimentation, instrumentalisation. ELA. Etudes de linguistique appliquee, 2(190), 195203. Litosseliti, L. (Ed.). (2018). Research methods in linguistics. Bloomsbury Publishing. Molinelli, P., & Putzu, I. (2015). Modelli epistemologici, metodologie della ricerca e qualità del dato. Dalla linguistica storica alla sociolinguistica storica. Franco Angeli. Péry-Woodley, M.-P., Afantenos, S. D., Ho-Dac, L.-M., & Asher, N. (2011). La ressource ANNODIS, un corpus enrichi d'annotations discursives. TAL, 52(3), 71101. Sarrica, M., Mingo, I., Mazzara, B., & Leone, G. (2016). The effects of lemmatization on textual analysis conducted with IRaMuTeQ: results in comparison. JADT2016: 13ème Journées Internacionales d'Analyse Statistique de Données Textuelles. Scaglione, F. (2018). "Lavorare"; il dato linguistico: Prospettive e limiti. Alcune considerazioni dall'esperienza dell'Atlante Linguistico della Sicilia (ALS). In G. Sampino (Éd.), Atti del convegno internazionale dei dottorandi (p. 101122). Tanguy, L. (2012). Complexification des données et des techniques en linguistique : contribution du TAL aux solutions et aux problèmes. HDR dissertation, Université de Toulouse 2 - le Mirail. Zalmout, N., Erdmann, A., & Habash, N. (2018). Noise-robust morphological disambiguation for dialectal Arabic. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) (pp. 953-964). Zeroual, I. (2018). Building Arabic Corpora: Concepts, Methodologies, Tools, and Experiments (Doctoral dissertation, University of Maryland, USA).

1 0