Analiza sentimenta novinskih članaka

master's thesis

• Title (english): News Sentiment Analysis
• Author: Ana Barić
• Mentor: Tomislav Burić (mentor)
• Committee member: Tomislav Burić (predsjednik povjerenstva)
• Committee member: Mervan Pašić (član povjerenstva)
• Committee member: Lenka Mihoković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-05, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Cilj ovog rada bio je razvoj modela za strojnu evaluaciju sentimenta u novinskim člancima. Tip sentimenta koji se izučavao bio je ciljani sentiment, odnosno sentiment prema ciljnom entitetu na razini rečenice. U sklopu razvoja modela za analizu sentimenta u ovom radu označen skup od 1636 podataka koji su bili korišteni za treniranje i evaluaciju modela. Za razvoj modela koristili su se pristupi iz domene dubokog učenja. Implementacija ovog rada obuhvaća razvoj cjevovoda za preprocesiranje... More podataka te implementaciju i evaluaciju unaprijednih i povratnih dubokih modela te podešavanje BERTić modela. Less

Abstract (english)

This paper aimed to develop a model for the machine evaluation of sentiment in newspaper articles. The type of sentiment studied was targeted sentiment, that is, sentiment toward the target entity at the sentence level. As part of developing the sentiment analysis model, a set of 1636 manually annotated data points was used to train and evaluate the model. Deep learning approaches were used to develop the model. The implementation of this paper includes the development of pipelines for... More data preprocessing and the implementation and evaluation of feed-forward and recurrent deep models and BERTić fine-tuning. Less

• Keywords: analiza sentimenta; duboki modeli; kontekstualizirane reprezentacije riječi; Bertic; TSA
• Keywords (english): sentiment analysis; deep learning; contextualized word embeddings; Bertic; TSA
• Language: croatian
• URN:NBN: urn:nbn:hr:168:441229
• 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 14:28:20