Sustav za izgradnju afektivnog rječnika za engleski jezik zasnovan na društvenim medijima

master's thesis

• Title (english): System for creating an affective lexicon for English language based on social media
• Author: Tea Pović
• Mentor: Vedran Podobnik (mentor)
• Committee member: Vedran Podobnik (predsjednik povjerenstva)
• Committee member: Ognjen Dobrijević (član povjerenstva)
• Committee member: Marina Bagić Babac (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

Razvojem Interneta, weba, a poglavito društvenih mreža iz korijena se promijenio način na
koji ljudi iznose svoja stajališta i mišljenja o širokom spektru tema. Samim time raste i
potreba za automatskim prepoznavanjem mišljenja, stavova ili emocija iz nestrukturiranog
teksta pisanog od strane ljudi, odnosno za strojnom analizom sentimenta pisanog teksta. U sklopu ovog rada korištenjem programskog jezika R i RStudio programskog sučelja ostvaren je sustav za automatizirano prepoznavanje emocija... More iz nestrukturiranog teksta napisanog na engleskom jeziku. Sustav omogućuje finu granulaciju teksta u pet kategorija: srdžba, humor, radost, tuga i iznenađenje. Osnova sustava je afektivni rječnik koji je izgrađen na temelju 52 968 statusa dohvaćenih s društvene mreže Facebook, a koji se zatim koristi za klasifikaciju novog teksta primjenom Naivnog Bayesovog algoritma. Razvijeni sustav postigao je vrlo dobre rezultate, koji nadmašuju rezultate puno složenijih sustava. Less

Abstract (english)

The development of the Internet, web and especially social networks has drastically changed the way people express their views and opinions on a variety of topics. Those views and opinions can be very important when it comes to Customer Experience management or social media analysis. Automatic sentiment analysis gives us the ability to process high volumes of data with minimum delay, high accuracy and low cost. In this master thesis, a system for the automated recognition of emotions from... More unstructured English text was built using the R programming language and the RStudio programming interface. The developed system can classify unstructured data into five categories: anger, haha, love, sad and wow. The most important part of the system is an affective lexicon that was built from 52 968 Facebook posts and applied to new data using a Naive Bayesian classifier. The developed system has achieved very good results, outperforming the results of much more complex systems. Less

• Keywords: analiza sentimenta; afektivni rječnik; engleski jezik; društveni mediji; Facebook; R; Naivni Bayesov algoritam klasifikacije; klasifikacija; regresija
• Keywords (english): sentiment analysis; emotion lexicon; English language; social media; Facebook; R; Naive Bayesian classifier; classification; regression
• Language: croatian
• URN:NBN: urn:nbn:hr:168:742786
• 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: 2019-03-14 00:09:41