Strojno učenje za analizu sentimenta u mikroblogovima

undergraduate thesis

• Title (english): Machine Learning for Sentiment Analysis in Microblogs
• Author: Ivan Križanić
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: Mladen Karan (član povjerenstva)
• Committee member: Martin Tutek (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-16, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj se rad bavi analizom sentimenta u mikroblogovima društvene platforme Twitter. Proučavanje teme analize sentimenta ostvareno je uz pomoć četvrtoga zadatka s natjecanja Semeval, koji je bio vrlo popularan zadatak nekoliko godina u nizu u kojima se natjecanje održavalo. Napravljena je usporedba pristupa klasičnim strojnim učenjem, odnosno korištenja SVM-modela i pristupa dubokog ućenja, odnosno korištenja povratne neuronske mreže (RNN) s arhitekturom ćelije s dugoročnom memorijom (LSTM). S pristupom klasičnog strojnog učenja ostvarena je točnost od 62.3%, a pristupom dubokog učenja ostvarena je nešto bolja točnost od 64.24%, što implementaciju stavlja na 10. mjesto implementacija koje su bile predane u sklopu natjecanja 2017. godine.

Abstract (english)

The theme of this thesis is sentiment analysis on micro blogs from the social platform Twitter. The study of sentiment analysis was done with a help of Semeval competition task 4, very popular and attractive task for several years in a row. Comparison of traditional machine learning and deep learning techniques was made by implementing two models, one using SVM, and the other one using RNN with LSTM architecture. The best result with SVM model was 62.3%, and 64.24% with LSTM. Implementation would take 10th place on scoreboard of Semeval 2017 edition of the competition.

• Keywords: strojno učenje; duboko učenje; obrada prirodnog jezika; analiza sentimenta; analiza mikroblogova; Semeval
• Keywords (english): Machine learning; Deep learning; Natural language processing; Sentiment analysis; Microblogs analysis; Semeval
• Language: croatian
• URN:NBN: urn:nbn:hr:168:044762
• 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: 2023-01-14 09:12:41