The Sentiment of Latin Poetry. Annotation and Automatic Analysis of the Odes of Horace

During the recent years, an always growing number of linguistic resources and automatic systems for sentiment analysis have been developed covering a wide range languages. However, research in this field is still not much explored texts written Classical Working on such languages means dealing with peculiar textual genres as philosophical, historical or religious treatises, epic narratives, plays poems. Poems are particularly suitable because they tell us about emotions passions. In paper, we describe creation first small gold standard Latin made poems by Horace manually annotated emotion polarity, but also report results set classification experiments. particular, test both lexicon-based approach, which uses polarity lexicon called LatinAffectus, zero-shot transfer method. We provide details methodology adopted annotation standard, development our experiments give limitations proposed approaches.