Analiza sentimenta zasnovana na modelima dubokog učenja

master's thesis

• Title (english): Sentiment Analysis Based On Deep Learning Models
• Author: Dunja Aćimović
• Mentor: Goran Delač (mentor)
• Committee member: Goran Delač (predsjednik povjerenstva)
• Committee member: Klemo Vladimir (član povjerenstva)
• Committee member: Marin Šilić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-04, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu su se detaljno opisale dvije različite arhitekture modela dubokog učenja: rekurzivne neuronske mreže i transformeri. Obje arhitekture su postigle dotad neviđene rezultate u području obrade prirodnog jezika, te su naglo postale popularne i pronašle mnogobrojne primjene, kao što je analiza sentimenta. Odabran je i implementirani po jedan model ovih arhitektura. Za primjer modela rekurzivnih neuronskih mreža odabran je model LSTM, dok je za primjer transformera odabran model BERT. Oba... More modela su se pokazala kao odličan izbor za analizu sentimenta. Less

Abstract (english)

This thesis explores two different deep learning arhitectures: recurrent neural networks and transformers. Both have shown state-of-the-art results on natural language processing tasks, so they have naturally became popular and found many applications, one of which is sentiment analysis. For each arhitecture, a model was chosen and implemented, it was LSTM for recurrent neural networks and BERT for transformers. Finally, both have shown capacity for exceptional results in the field of... More sentiment analysis.
Less

• Keywords: analiza sentimenta; rekurzivne neuronske mreže; LSTM; transformeri; BERT; duboko učenje
• Keywords (english): sentiment analysis; recurrent neural networks; LSTM; transformers; BERT; deep learning
• Language: croatian
• URN:NBN: urn:nbn:hr:168:917890
• 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: 2023-01-13 11:39:15