Određivanje stava iz objava na Twitteru pomoću strojnog učenja

undergraduate thesis

• Title (english): Detecting stance in tweets using machine learning
• Author: Mislav Perić
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: Zoran Medić (član povjerenstva)
• Committee member: Domagoj Pluščec (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-01, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Određivanje stava je problem unutar područja obrade prirodnog jezika koji se u posljednje vrijeme počeo sve više istraživati. U ovom je radu opisan i implementiran sustav određivanja stava iz objava na društvenoj mreži Twitter. Uspostavljen je osnovni model SVM te modeli dubokog učenja temeljeni na predtreniranom modelu BERT. Uspoređeni su različiti duboki modeli koji koriste višezadaćno učenje s pomoćnim zadatcima određivanja sentimenta i subjektivnosti. Također je ispitana uloga učenja ansambla.

Abstract (english)

Stance detection is a problem within the field of natural language processing that has recently begun to be increasingly researched. This thesis describes and implements a system for detecting stances from posts on the social network Twitter. The baseline SVM and deep learning models based on the pre-trained BERT model have been established. Different deep learning models that use multitasking learning with auxiliary tasks of determining sentiment and subjectivity were compared. The role of ensemble learning was also examined.

• Keywords: obrada prirodnog jezika; određivanje stava; strojno učenje; duboko učenje; višezadaćno učenje; učenje ansambla
• Keywords (english): natural language processing; stance detection; machine learning; deep learning; multitask learning; ensemble learning
• Language: croatian
• URN:NBN: urn:nbn:hr:168:606194
• 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: 2023-01-15 11:02:08