Otkrivanje emocija u digitalnom sadržaju

undergraduate thesis

• Title (english): Emotion detection in digital content
• Author: Sanja Hanić
• Mentor: Dragan Jevtić (mentor)
• Committee member: Dragan Jevtić (predsjednik povjerenstva)
• Committee member: Renato Šoić (član povjerenstva)
• Committee member: Ivan Bartolec (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Ovaj rad bavi se otkrivanjem emocija u tekstu te je napravljen kratak pregled dosadašnjih metoda za takvu obradu teksta. Opisana je jedna neuronska mreža s različitim parametrima koja radi predikciju emocije u kratkom tekstualnom zapisu - tweetu. Predikcija emocije je zapravo klasificiranje teksta kao jedne od šest mogućih emocija. Računalna obrada teksta je složen zadatak, no različite metode strojnog učenja imaju obećavajuće rezultate. Uz strojno učenje istaknuti su i drugi bitni procesi obrade teksta, poput tokenizacije i vektorske reprezentacije riječi.

Abstract (english)

This paper is focused on emotion detection in textual content and it represents a brief summary of basic emotion detection methods. The neural network described in this paper is used to predict an emotion of a short statement or a tweet. Emotion prediction is actually a classification problem, where every tweet is classified as one of the six possible emotions. Computer language processing is a complex task but machine learning methods are showing promising results. Alongside machine learning, this paper presents other important parts of natural language processing, like word tokenization and word embedding.

• Keywords: detekcija emocija; obrada prirodnog jezika; tokenizacija riječi; strojno učenje; višeklasna klasifikacija; neuronske mreže
• Keywords (english): emotion detection; natural language processing; word tokenization; machine learning; multiclass classification; neural networks; word embedding
• Language: croatian
• URN:NBN: urn:nbn:hr:168:971789
• 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: 2023-01-13 15:03:16