Lev Shadrin · Zenodo (CERN European Organization for Nuclear Research) 2026 · 2026
DOI: 10.5281/zenodo.22896152
Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).
A poster presented at the MLDSE workshop in Graz (21-25 September, 2026), prepared as part of the LAGOOS (ΛΑΓΩΌΣ) project, funded by the Austrian Science Fund (FWF, Grant DOI: 10.55776/Y1519) and hosted by the University of Innsbruck. Key highlights: HTR model training progress CER improvement (LAGOOS HTR v.9, CER = 4.56%) using custom trained Transkribus Fields Models to improve diacritics identification and tachygraphic abbreviation expansion for the Greek scholarly hand TF-IDF-based topic modelling to identify salient vocabulary in the entries verb frequency analysis to trace the use patterns of different historical registers of Greek in the diary
No comments yet — start the discussion below.