1The work described in this paper was funded by CLARIN.SI – the Slovene national node of the European Common Language Resources and Technology Infrastructure CLARIN, the ARRS P6-0215 research program (Slovene language – basic, contrastive, and applied studies) and the H2020 project SSHOC – Social Sciences & Humanities Open Cloud (GA No. 823782). We would also like to thank Prof. Dr. Milica Antić Gaber, Assist. Prof. Dr. Irena Selišnik, Prof. Dr. Cornelia Ilie, Tjaša Cankar, Mladen Zobec and Ruben Ros for their thoughtful comments.