Postupci analize sentimenta kroz domene za hrvatski jezik

undergraduate thesis

• Title (english): Cross-Domain Sentiment Analysis Methods for Croatian Language
• Author: Vinko Kašljević
• 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: 2018-09-05, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Danas postoji trend sve većeg korištenja strojnog učenja za različite probleme. Jedan od tih problema je obrada prirodnog jezika gdje strojno učenje postiže jako dobre rezultate. Da bi mogli koristiti strojno učenje potrebni su nam podaci te iako podataka imamo više nego ikada prije, nama su potrebi dobro označeni podaci. Tu nam može pomoći prilagodba domene.

U ovom radu objašnjavamo što je strojno učenje, što je prilagodba domene te kako nam ona može pomoći pri analizi sentimenta za različite... More domene. Zatim razmatramo praktični dio ovog rada: priprema podataka, odabir modela te izvođenje različitih eksperimenata na kojima vidimo koliko dobro radi prilagodba domene nad više domena.
Less

Abstract (english)

Today there is a trend of ever increasing use of machine learning for different problems. One of these problems is natural language processing where machine learning achieves very good results. In order to be able to use machine learning, we need data, and although we have more data than ever before, we need well-labeled data. Domain adaptation can help us with this problem.

In this paper, we explain what is machine learning, which is domain adaptation, and how it can help us in the sentiment... More analysis for different domains. Then we consider the practical part of this work: data preparation, model selection and performing different experiments to see how well does domain adaptation work.
Less

• Keywords: strojno učenje; obrada prirodnog jezika; analiza sentimenta; stroj potpornih vektora; prilagodba domene
• Keywords (english): machine learning; natural language processing; sentiment analysis; support vector machine; domain adaptation
• Language: croatian
• URN:NBN: urn:nbn:hr:168:457933
• 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 21:27:48