Analiza sentimenta u paralelnim korpusima hrvatskog i poljskog jezika

undergraduate thesis

• Title (english): Sentiment analysis in parallel corpora of the Croatian and Polish languages
• Author: Dominik Vrbanec
• Mentor: Nives Mikelić Preradović (mentor)
• Committee member: Nives Mikelić Preradović (član povjerenstva)
• Committee member: Sanja Kišiček (predsjednik povjerenstva)
• Committee member: Tomislava Lauc (član povjerenstva)
• Granter: University of Zagreb; Faculty of Humanities and Social Sciences; (Department of information and Communication sciences); Zagreb
• Defense date and country: 2023-07-04, Croatia
• Scientific / art field, discipline and subdiscipline: SOCIAL SCIENCES; Information and Communication Sciences

Abstract

U području obrade jezika postoji nedostatak hrvatskog korpusa teksta označenog sa sentimentima koji omogućuje njihovu analizu. U radu je opisan proces stvaranja dva takva korpusa, tj. paralelnog hrvatsko-poljskog korpusa, a na temelju postojećih oznaka iz engleskog korpusa. Engleski izvornik je nastao na temelju televizijske serije Friends. Strojnom obradom u vlastitom Python programu, prevedeni su na hrvatski i poljski jezik tekstovi iz engleskog korpusa te izvršena njihova analiza. U radu su prikazani i dokumentirani problemi s kojima se je pri tome trebalo suočiti i koje je trebalo riješiti.

Abstract (english)

In the field of natural language processing, there is a lack of a Croatian text corpus marked with sentiments, which enables their analysis. This paper describes the process of creating two such corpora, i.e., a parallel Croatian-Polish corpus, based on existing tags from the English corpus. The English original was created based on the TV Show Friends. Using machine processing, i.e., translation via the original Python program, texts from the English corpus were translated into Croatian and Polish equivalents and their analysis was carried out. The paper presents and documents the problems that had to be faced and solved during that process.

• Keywords: paralelni korpus; sentiment; strojna obrada jezika
• Keywords (english): parallel corpus; sentiment; machine language processing
• Language: croatian
• URN:NBN: urn:nbn:hr:131:466759
• Study programme: Title: Information Sciences; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) informacijskih znanosti (sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) informacijskih znanosti)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2023-07-04 19:12:55