Naučno-popularno predavanje: Determining sentiment of tweets using the lexicon and (AnA)-affirmative and non-affirmative words

Sentiment Analysis, is also known as Opinion Mining, is a field within Natural Language Processing (NLP) that builds systems that try to identify and extract opinions within the text.These systems extract attributes of the expression e.g.: Subject, Opinion (positive or negative) and Opinion holder.
Sentiment analysis gives us the opportunity to transform unstructured information in structured data, that later can be useful for commercial applications: marketing analysis, public relations, product reviews …The talk will present a simple model for sentiment tagging of Tweeter messages based on AnA (affirmative and non-affirmative) wordsand tagged sentiment dictionary (Kadunc-Šikonja dictionary).Lexicon was used in combination with intensifiers, where intensifier (AnA word),that stand next to the word, produce more powerful (more positive or more negative) sentiment value of word – tweet at all. This also presents our first step in creating the first Bosnian lexicon and proceed the same on Sentiment analysis of tweets written in Bosnian where the major topic will be tweeting about Bosnian National basketball players: Džanan Musa and Jusuf Nurkić.


Predavanje će se održati u četvrtak 12.03.2020 od 14:oo u učionici 205 a predavač je Sead Jahić, student trećeg ciklusa na   Odsjeku za informacione nauke i tehnologije na Univerza na Primorskem, Slovenija