Towards Flexible Cross-Resource Exploitation of Heterogeneous Language Documentation Data

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 2901–2905 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 2901 Towards Flexible Cross-Resource Exploitation of Heterogenous Language Documentation Data Daniel Jettka, Timm Lehmberg Universität Hamburg, Institut für Finnougristik/Uralistik Max-Brauer-Allee 60, D-22765 Hamburg {daniel.jettka, timm.lehmberg}@uni-hamburg.de Abstract This paper reports on challenges and solution approaches in the development of methods for language resource overarching data analysis in the field of language documentation. It is based on the successful outcomes of the initial phase of an 18 year long-term project on lesser resourced and mostly endangered indigenous languages from the Northern Eurasian area, which included the finalization and publication of multiple language corpora and additional language resources. While aiming at comprehensive cross- resource data analysis, the project is simultaneously confronted with a dynamic and complex resource landscape, that especially results from a vast amount of multi-layered information stored in the form of analogue primary data in different widespread archives on the territory of the Russian Federation. The described methods aim at solving the tension between the needs for unification of heterogenous data sets and vocabularies on the one hand and maximum openness for the integration of future resources and the adaption of external information on the other hand. Keywords: Digital infrastructure, Language Documentation, Linked Data, Knowledge Graphs 1. Introduction Since its beginning in 2016 and over an expected runtime of 18 years the INEL project („Grammatical Descriptions, Corpora and Language Technology for Indigenous Northern Eurasian Languages“)1 aims at the discovery and documentation of language material for selected lesser documented and mostly endangered indigenous languages and varieties from the Northern Eurasian Area. In several subprojects, each focussing on one particular language, these resources are successively curated with digital methods, enriched with linguistic information, and are finally published in the form of linguistic corpora and other types of language resources. In association with this work comprehensíve linguistic description and analysis is executed over the entire runtime period and on all levels of language descriptions, also within satellite projects. According to its extended areal, temporal, and also disciplinary focus the aims of INEL go beyond those of many other research projects in the field of language documentation. While the large-scale approach of the project allows for manifold cross-lingual, diachronic and interdisciplinary research on a very high level, the data situation that will be described in the following section puts new requirements to methods of data modelling and technology used within the project. Thereby, major challenges arise from the need for cross-resource data analysis and a rather complex resource landscape on the territory of the Russian Federation. 2. Language Resources and Digital Infrastructure 2.1 Complex Heterogeneous Information Sources The amount of more or less structured collections of object and meta language data resulting from research on lesser documented languages all over the world is incalculable per se. In particular with regard to indigenous languages 1 https://inel.corpora.uni-hamburg.de/ from the Northern Eurasian area a lot of research material has been created …

Content truncated to 500 words