Modeli strojnog učenja za zadatak razumijevanja narativnog teksta

undergraduate thesis

• Title (english): Machine Learning Models for Narrative Text Understanding
• Author: Ivan Crnomarković
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: Mladen Karan (član povjerenstva)
• Committee member: Domagoj Alagić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Zahvaljujući internetu i dinamičnosti današnjeg društva ljudi generiraju dosad
neviđene količine podataka. Kako je velik dio generiranih podataka u nestrukturiranom
tekstnom obliku, radi se na sustavima koji bi mogli probaviti ove
masivne tekstove i razumjeti njihovo značenje. Jedan od načina testiranja takvih
sustava jest test dovršavanja priče, tj. test u kojem sustav na osnovi priče od
četiri rečenice pokušava pogoditi koji je od dvaju ponuđenih završetaka točan.
Danas najbolje rezultate na... More testu dovršavanja priče postižu sustavi temeljeni na
metodama strojnog i dubokog učenja. Stoga smo implementirali dva modela temeljena
na metodama strojnog i dubokog učenja, kako bi dali pregled postupaka
implementacije tih metoda na rješavanju testa završetka priče. Less

Abstract (english)

Thanks to the internet and the dynamics of today’s society, people are generating
unprecedented amounts of data. Since much of the generated data is in an
unstructured text form, we are developing systems that could digest these massive
texts and understand their meaning. One of the ways to test such systems is
a story cloze test, i.e. a test in which a system reads 4-sentence story and based
on that tries to guess which of the two offered endings is correct. Today, the
best results on the... More story cloze test are achieved by the systems based on machine
and deep learning methods. So we implemented two models based on machine
learning and deep learning methods to review the implementation procedures of
these methods on task of solving a story cloze test. Less

• Keywords: strojno učenje; duboko učenje; word2vec; narativni tekst; procesiranje prirodnog jezika; LSDSem 2017; test završetka priče
• Keywords (english): machine learning; deep learning; word2vec; narrative text; natural langugage processing; LSDSem 2017; story cloze test
• Language: croatian
• URN:NBN: urn:nbn:hr:168:853403
• 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: 2019-03-13 19:37:39