The contents of this repository is the result of virtual mobility grant (01/08/2022 to 15/10/2022) within NexusLingurum COST Action CA CA18209. The main goal was to publish 300 novels from ELTeC - European Literary Text Collection from period 1840-1920., 100 per 3 languages (Serbian, Slovenian adn Portugese) as open linked data according to best practice and guidelines fostered by NexusLinguarum.
Namely, ELTeC novels in so-called level-2 format were developed within COST Action CA16204 Distant Reading for European Literary History (2017 - 2022) and they are valuable resource to be published in LLOD. Three data collection were used: https://github.com/COST-ELTeC/ELTeC-por/tree/master/level2, https://github.com/COST-ELTeC/ELTeC-slv/tree/master/level2, https://github.com/COST-ELTeC/ELTeC-srp/tree/master/level2.
Metadata for selected novels are linked with already available in Wikidata, named WikiELTeC, as described in "From ELTeC Text Collection Metadata and Named Entities to Linked-data (and Back)" by Milica Ikonic Nešić, Ranka Stanković, Christof Schoch, Mihailo Škorić, http://www.lrec-conf.org/proceedings/lrec2022/workshops/LDL/pdf/2022.ldl2022-1.2.pdf
The NLP Interchange Format (NIF), designed to facilitate the integration of NLP tools in knowledge extraction pipelines, provides support for part-of-speech tagging, lemmatization and entity annotation, enabling ELTeC level-2 layers transformation. Python code, including Jupyter notebook, is prepared for export from XML/TEI into NIF, available in colab subfolder. For Wikidata management mkwikidata (https://pypi.org/project/mkwikidata/) library was used and for working with RDF rdflib (https://rdflib.readthedocs.io/en/stable/index.html).
Samples of NIF files are in NIF subfolder.
In March 2023 the approach was improved, 9 languages are included. New jupiter notebook is available in folder notebook-9-lngs and data are in http://llod.jerteh.rs/ELTEC/. Related paper ...