Model analize sentimenta temeljen na n-gramima

master's thesis

• Title (english): N-gram based sentiment analysis model
• Author: Fran Leontić
• Mentor: Marko Horvat (mentor)
• Committee member: Marko Horvat (predsjednik povjerenstva)
• Committee member: Mirjana Domazet-Lošo (član povjerenstva)
• Committee member: Tomislav Jagušt (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-07, Croatia

Abstract

Diplomski rad bavi se primjenom analize sentimenta i emocija na tekstove na hrvatskom jeziku koristeći pristup temeljen na n-gramima, tj. pristup temeljen na leksikonu. U sklopu rada izrađeni su modeli za ocjenjivanje tekstova koji se temelje na dimenzijskom modelu emocija i diskretnom modelu emocija. Ocijenjeni su tekstualni dokumenti od kasne 2019. godine do početka 2021. godine, koji su značajni jer predstavljaju razdoblje prva dva vala COVID-19 pandemije u Hrvatskoj. Iz ocijenjenih tekstova pomoću dimenzijskog modela zaključeno je da je pandemija izazvala neugodu, da su emocije bile niskog intenziteta i da ljudi nisu imali osjećaj kontrole nad emocijama. Iz ocijenjenih tekstova pomoću diskretnog modela zaključeno je da su prevladavajuće emocije strah i tuga. Izrađena je web-aplikacija koja demonstrira primjenu navedenih modela na proizvoljne tekstove.

Abstract (english)

This thesis focuses on the application of sentiment analysis and emotion analysis on Croatian texts using an n-gram or lexicon based approach. The developed models for text evaluation are based on the dimensional and discrete emotion models. Text documents from late 2019. to early 2021. were evaluated. They are significant because they represent the first two COVID-19 pandemic waves in Croatia. The dimensional model's ratings show that the pandemic caused discomfort, that the emotions were of low intensity and that the people did not have control over their emotions. The discrete model's ratings show that the prevailing emotions were fear and sadness. A web-application that demonstrates the usage of the mentioned models on user inputs was developed.

• Keywords: analiza sentimenta; analiza emocija; COVID-19; n-gram; leksikon
• Keywords (english): sentiment analysis; emotion analysis; COVID-19; n-gram; lexicon
• Language: croatian
• URN:NBN: urn:nbn:hr:168:171027
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-10-11 09:32:06