Primjena modela dubokog učenja za otkrivanje stavova u korisničkim komentarima

undergraduate thesis

• Title (english): Application of Deep Learning for Stance Detection in User Comments
• Author: Bartol Freškura
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: Domagoj Alagić (član povjerenstva)
• Committee member: Martin Tutek (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2016-07-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Duboko učenje je relativno nova grana strojnog učenja koja se temelji na teoriji
neuronskih mreža. Pomoću naprednih arhitektura dubokih neuronskih mreža danas
se pokušavaju riješiti poznati problemi u području obrade prirodnog jezika kao što su
analiza stavova i sentimenta. U sklopu završnog rada napravljene su dvije poznate
konfiguracije povratnih neuronskih mreža LSTM i GRU pomoću programa Tensor-
flow. Provedeni su razni eksperimenti u kojima se traže optimalni parametri povratnih
neuronskih mreža s ciljem točne klasifikacije stavova u korisničkim komentarima iz
skupova podataka na hrvatskom i engleskom jeziku.

Abstract (english)

Deep learning is a relatively new branch of machine learning based on the theory of
neural networks. Using the advanced deep neural network architectures, today’s well-
known natural language processing problems like sentiment and opinion mining are
being solved. Two widely used configurations of recurrent neural networks were cre-
ated with the Tensorflow software: LSTM and GRU. Many different experiments were
conducted in which the focus was on finding the optimal parameters of the recurrent
neural networks, which would yield in correct stance classification of user comments
from the Croatian and English datasets.

• Keywords: umjetna neuronska mreža; duboko učenje; LSTM; GRU; strojno učenje; obrada prirodnog jezika
• Keywords (english): artificial neural network; deep learning; LSTM; GRU; machine learning; natural language processing
• Language: croatian
• URN:NBN: urn:nbn:hr:168:883531
• 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 08:46:55