Accueil > Thèmes de recherche > Uralic Linguistics

Research on Uralic Languages at LATTICE

publié le , mis à jour le

Presentation

LATTICE has always had a close relationship with Uralic languages, originally thanks to researchers like Marie-Josèphe Gouesse, Peter Balog or Eva Havu. This line of research is still ongoing at LATTICE, along with focused international collaborations.

We are investigating purely linguistic issues as well as computational ones. On the one hand, we are interested in the syntax-semantic interface (e.g. noun phrase structure, verb subcategorization, aspect and differential object marking) and on the other hand on various aspects of language processing (lexical acquisition, tagging and parsing). We focus mainly on Finnish, Saami, Komi and Udmurt.

Funded Projects

  • LAKME is a PSL funded project (2016-2018) exploring new NLP techniques (esp. machine learning techniques) to annotate scholarly relevant corpora. The project focuses on morphologically-rich languages that are especially challenging for current NLP systems. Three languages (or groups of languages) are considered : Rabbinic Hebrew, Medieval French and some Uralic languages (esp. Finnish, Komi and Udmurt). The project is a collaboration between Lattice (PI, Thierry Poibeau), the Ecole Pratique des Hautes Etudes (Daniel Stoekl Ben Ezra) and the Ecole Nationale des Chartes (Jean-Baptiste Camps). The project is thus not restricted to Uralic languages but we expect to be able to re-use techniques that have proven useful for other morphologically-rich languages.

We are also collaborating with The Institute Of Linguistics of the Russian State University for the Humanities, thanks to a RGNF-CNRS project funded by the two institutions. The goal of this project is to propose a multimodal analysis of the noun phrase in Finno-Ugric languages, with a specific focus on referential continuity and on information structure of the sentence. The project PI are Benjamin Fagard for LATTICE and Natalia Serdobolskaya for the Russian State University for the Humanities. Computational approaches are also explored within this framework, in particular with the National Research University, Higher School of Economics, with Svetlana Toldova and
Timofey Arkhangelskiy.

Invited Foreign Researchers

Most researchers have been invited thanks to ENS/PSL and more specifically the support of TransferS (laboratoire d’excellence, program “Investissements d’avenir” ANR-10-IDEX-0001-02 PSL* and ANR-10-LABX-0099).

International Collaborations

We also have relations with other groups, especially at the University of Helsinki (Janne Saarikivi).

National Collaborations

We have close relationships with the Institut National des Langues et Civilisations Orientales INaLCO where one can find the best specialists on the topic in France. The 2016 workshop on the Saami language and culture was for a part organized at the Ecole normale supérieure.

Online demonstrations

An interface to explore which Finnish verbs (mainly) subcategorize partitive objects and which ones mainly subcategorize total objects (i.e. nominative, accusative and genitive) is available HERE. It is based on the analysis of the Finnish part of the Europarl corpus. Caution : The file is very heavy and the interface is not optimized.