Sentiment Analysis (SA) of the Bosnian Language (the overview of contributions to SA in Bosnian)
On the seminar, the contributions to Sentiment Analysis in the Bosnian language will be presented. Insights into the lexicon creation process will be provided first, followed by the discussion of ideas for annotating and ensuring lexicon coverage. The speaker will introduce the MERI (Modifying Emotion-related Intensity) equation for sentiment calculation and demonstrate how Negation clues and AnAwords (Affirmative and non-affirmative) words influence the overall sentiment value of the text through two models they developed: BoSa and BOSentiment. Additionally, a comparison of these models with other ML techniques and CNN will be presented.
ABOUT THE PRESENTER:
Sead Jahić is in the final year of his PhD. Prior to that, he pursued mathematics at the University of Tuzla (UNTZ) in Bosnia and Hercegovina. In terms of work experience, he served as a teaching assistant at UNTZ in the field of Applied Mathematics and Computer Science. Currently, he is a Teaching Assistant at UP FAMNIT, focusing on Java from introductory levels through GUI to Parallel/Concurrent and Distributed programming.