Aplikacija za određivanje sentimenata u tekstu korištenjem ugrađenih vektora značajki

undergraduate thesis

• Title (english): Application for text sentiment analysis using embedded feature vectors
• Author: Ariana Dobrostal
• Mentor: Marko Horvat (mentor)
• Committee member: Marko Horvat (predsjednik povjerenstva)
• Committee member: Tomislav Jagušt (član povjerenstva)
• Committee member: Juraj Dončević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-02-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad koristi Word2Vec model za analizu sentimenata nad tri skupa podataka sastavljenih od objava s društvene mreže Twitter prikupljenih u razdoblju neposredno prije, tijekom i neposredno poslije prvog vala pandemije COVID-19 virusa, odnosno u intervalu od kraja 2019. godine do sredine 2020. godine. Word2Vec model je treniran podatcima iz sva tri skupa koji ne pripadaju skupu za analizu, a oznake sentimenata su dobivene pomoću VADER-a. U analizi promatramo promjene udjela pozitivnih, neutralnih i negativnih podataka, a zatim odabiremo najbolji skup podataka te nad njim izvršavamo detaljniju analiza povećavanjem intervala i broja trenutaka promatranja.

Abstract (english)

This paper uses the Word2Vec model to analyze sentiments over three sets of data composed of posts from the social network Twitter collected in the period immediately before, during and immediately after the first wave of the COVID-19 virus pandemic, i.e. in the interval from the end of 2019 to the middle of 2020 . The Word2Vec model was trained with data from all three sets that do not belong to the analysis set, and the sentiment labels were obtained using VADER. In the analysis, we observe changes of the percentages of positive, neutral and negative data, then select the best set of data and perform a more detailed analysis on it by increasing the interval and number of observation moments.

• Keywords: Word2Vec; analiza sentimenata; ugradbene matrice riječi; COVID-19; Twitter; VADER
• Keywords (english): Word2Vec; sentiment analysis; word embeddings; COVID-19; Twitter; VADER
• Language: croatian
• URN:NBN: urn:nbn:hr:168:728903
• 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: 2023-07-13 10:01:23