Research projects

Transforming archival research on the Holocaust, by providing methods and tools to integrate and provide access to a wide variety of archival content.

Coordinating Digital Humanities infrastructure activities in Europe (17 partners, 5 associated partners).

In order to support the fight against radicalization and thus prevent future terrorist attacks from taking place, the CounteR project brings data from disperse sources into an analysis and early alert platform for data mining and prediction of critical areas (e.g. communities), aiming to be a frontline community policing tool which looks at the community and its related risk factors rather than targeting and monitoring individuals. The system will incorporate state of the art NLP technologies combined with expert knowledge into the psychology of radicalization processes to provide a complete solution for law enforcement authorities to understand the when, where and why of radicalization in the community.

This project aims at building a cultural history of Artificial Intelligence, based on a mixed method, combining the methods of the history of ideas and the history of collective imaginations with a search of scientific literature and ethnographic field work among AI creators.

The MaTOS (Machine Translation for Open Science) project aims to develop new methods for the machine translation (MT) of complete scientific documents, as well as automatic metrics to evaluate the quality of these translations. Our main application target is the translation of scientific articles between French and English, where linguistic resources can be exploited to obtain more reliable translations, both for publication purposes and for gisting and text mining. However, efforts to improve MT of complete documents are hampered by the inability of existing automatic metrics to detect weaknesses in the systems and to identify the best ways to remedy them. The MaTOS project aims to address both of these issues.

The PRAIRIE Institute (PaRis AI Research InstitutE) is one of the four French Institutes of Artificial Intelligence, which were created as part of the national French initiative on AI announced by President Emmanuel Macron on May 29, 2018. PRAIRIE’s objective is to become within five years a world leader in AI research and higher education, with an undeniable impact on economy and technology at the French, European and global levels. It brings together academic members (“PRAIRIE chairs”) who excel at research and education in both the core methodological areas and the interdisciplinary aspects of AI, and industrial members that are major actors in AI at the global level and a very strong group of international partners.

The text simplification algorithm developed within Cap'FALC is based on neural models for natural language processing. It will work similarly to a spell checker, which marks passages in a text, offers solutions but does not correct without a human validation step. The tool is intended to represent a valuable aid for disabled people responsible for transcribing texts in FALC, not to replace their intervention at all stages of the drafting; only their expertise can validate a text as being accessible and easy to read and understand.

Empirical foundations of linguistics, including computational linguistics and natural language processing. ALMAnaCH’s predecessor team ALPAGE was one of the partner teams of this LabEx, which gathers a dozen of teams within and around Paris whose research interests include one aspects of linguistics or more. Several ALMAnaCH members are now “individual members” of the LabEx EFL. B. Sagot serves as deputy head (and former head) of one of the scientific strands of the LabEx, namely strand 6 dedicated to language resources. Benoît Sagot and D; Seddah are (co-)heads of a number of scientific “operations” within strands 6, 5 (“computational semantic analysis”) and 2 (“experimental grammar”). Main collaborations are related to language resource development (strands 5 and 6), syntactic and semantic parsing (strand 5, especially with LIPN [CNRS and U. Paris 13]) and computational morphology (strands 2 and 6, especially with CRLAO [CNRS and Inalco] and LLF [CNRS and Paris-Diderot]).

ALMAnaCH is a member of the CORLI consortium on “corpora, languages and interactions” (B. Sagot is a member of the consortium’s board).

The project is intended to digitise and exploit the early editions (beginning of the 20th century) of the Petit Larousse dictionary. ALMAnaCH is involved in the automatic extraction of the dictionary content by means of the GROBID-dictionary and in defining a TEI-compliant interchange format for all results.

The aim of the project is to make cancer data from health institutions accessible to all stakeholders involved for research and innovation purposes. The data at hand will be standardised and structured, in particular by extracting information from textual documents.

The HTRomance project is based on handwriting recognition (HTR). In particular, it proposes to evaluate and improve the capabilities of this technology when applied to literary manuscripts and public and private archives, in Latin and Romance languages, from the 11th to the 19th century, kept at the French National Library. The main objective of the project is the production of training data and transcription models resistant to changes in handwriting and language. It also intends to produce language models applicable to documents in ancient languages or to ancient language states. The development of training corpora will be accompanied and consolidated by the development and implementation of a novel process for evaluating the readability of output texts and the costs of producing new training data for HTR. HTRomance is complementary to editing or data mining projects: the models produced are likely to be used to obtain the textual data needed for editing or text mining.

The ambition of the ExcellencES TIERED project is to address the challenges of democratic systems in the face of environmental transformations and the digital transition, by producing outstanding scientific research, disseminating it within society, and training today's and tomorrow's decision-makers.

The Universal Dependencies project is an open community effort with over 300 contributors producing nearly 200 treebanks in over 100 languages. Universal Dependencies is a framework for consistent annotation of grammar (parts of speech, morphological features, and syntactic dependencies) across different human languages. With a release every 6 months since 2017, the UD framework is the de-facto standard for syntactic structures representations and of course the basis for most if not all supervised neural parsers.