Research project Variation in the texts of Georg Mancelius

A sub-corpus of Mancelius' texts published during his lifetime in the 17th century will be developed and explored for the study of (mainly) morphological variants. NLP tools will be applied to Old Latvian texts. Taking into account certain phonetic and morphological deviations from Modern Standard Latvian, it is planned to evaluate, adapt, and use an existing POS tagger (via Latvian NLP Pipeline as a Service (LV-PIPE https://nlp.ailab.lv/). Morphological annotation will first be carried out entirely manually (for some 50 000 running words). Later on, the training will be completed, and the annotation will be done semi-automatically with manual post-editing. Analysis of morphological variants will be carried out, detecting main types.