Kontekstno ovisna analiza sentimenta izraza hrvatskoga jezika

undergraduate thesis

• Title (english): Contextual Sentiment Analysis of Croatian Expressions
• Author: Paula Gombar
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: Goran Glavaš (član povjerenstva)
• Committee member: Domagoj Alagić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Porastom raspoloživih količina korisnički generiranog sadržaja povećalo se zanimanje za strojnom analizom sentimenta, kojom se utvrđuje je li tekst usmjeren pozitivno, negativno ili neutralno. U radu su proučeni postupci za analizu sentimenta s naglaskom na postupak temeljen na strojnom učenju. Razrađen je model za kontekstno ovisnu analizu sentimenta izraza hrvatskoga jezika temeljen na modelu nadziranog strojnog učenja, po uzoru na rad Wilsona i dr. (2005). Izgrađen je i ručno označen odgovarajući skup tekstnih podataka na hrvatskome jeziku za razvoj i ispitivanje modela. Provedeno je iscrpno eksperimentalno vrednovanje modela, statistička obrada rezultata i analiza pogrešaka.

Abstract (english)

Given the increase in the amount of available user-generated content, there has been rising interest in sentiment analysis based on machine learning, which determines whether the text attitude is positive, negative or neutral. This paper examines methods for sentiment analysis with a special focus on methods based on machine learning. A model for contextual sentiment analysis of Croatian expressions has been devised, based on the supervised machine learning model described in the work of Wilson et al. (2005). An appropriate dataset consisting of texts in Croatian was manually built and annotated. It was used for model development and testing. An exhaustive experimental evaluation of the model was conducted, along with statistical result and error analysis.

• Keywords: obrada prirodnog jezika; analiza sentimenta; strojno učenje; stroj potpornih vektora; hrvatski jezik; računalna lingvistika
• Keywords (english): natural language processing; sentiment analysis; machine learning; support vector machines; Croatian language; computational linguistics
• Language: croatian
• URN:NBN: urn:nbn:hr:168:839815
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2018-02-06 08:56:47