Analiza mnijenja u podatkovnom skupu SentiGMapsCro

master's thesis

• Title (english): Sentiment analysis in the SentiGMapsCro data set
• Author: Brigita Gašparić
• Mentor: Nives Mikelić Preradović (mentor)
• Committee member: Tomislava Lauc (predsjednik povjerenstva)
• Committee member: Sanja Kišiček (član povjerenstva)
• Committee member: Nives Mikelić Preradović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Humanities and Social Sciences; (Department of information and Communication sciences); Zagreb
• Defense date and country: 2024-09-16, Croatia
• Scientific / art field, discipline and subdiscipline: SOCIAL SCIENCES; Information and Communication Sciences; Organization and Informatics

Abstract

Ovaj diplomski rad istražuje analizu sentimenta na temelju aspekta unutar podatkovnog skupa SentiGMapsCro koji je nastao sakupljanjem najbolje i najgore recenziranih mjesta i objekata na području Grada Zagreba u Hrvatskoj s internetske stranice https://www.top-rated.online. Jezik, temeljni element ljudske komunikacije, prenosi informacije, emocije i mnijenje. Ovaj rad naglašava važnost analize sentimenta u razumijevanju subjektivnih iskustava podijeljenih na Internetu. Opisuje se postupak ručnog označavanja podatkovnog skupa SentiGMapsCro unutar alata za označavanje pod nazivom INCEpTION, s oznakama sentimenata u rasponu od -2 (vrlo negativan sentiment) do +2 (vrlo pozitivan sentiment) za obuku modela strojnog učenja. Rad naglašava izazove u analizi sentimenta, kao što su složenost prirodnog jezika, prisutnost ironije i metafora, problemi segmentacije te gramatičke pogreške. U radu se ocrtavaju teoretski okviri i istraživačke metodologije korištene u procesu označavanja, ilustrirajući poteškoće s kojima se istraživači susreću u računalnoj obradi prirodnog jezika.

Abstract (english)

The thesis explores sentiment analysis based on aspect within the SentiGMapsCro dataset, focusing on the best and worst reviews of places in Zagreb, Croatia extracted from https://www.top-rated.online. Language, a fundamental element of human communication, conveys information, emotions, and opinions. The study emphasizes the importance of sentiment analysis in understanding subjective experiences shared online. It describes the manual labeling process of the dataset within the annotation tool called INCEpTION, with sentiments ranging from -2 (very negative sentiment) to +2 (very positive sentiment), to train machine learning models. The thesis highlights challenges in sentiment analysis, such as natural language complexities and the presence of irony, metaphors, segmentation issues, and grammatical errors. It outlines theoretical frameworks and research methodologies employed in the labeling process, illustrating the difficulties encountered in computer-based natural language processing.

• Keywords: analiza sentimenta; podatkovni skup; SentiGMapsCro; INCEpTION
• Keywords (english): sentiment analysis; dataset; SentiGMapsCro; INCEpTION
• Language: croatian
• URN:NBN: urn:nbn:hr:131:991862
• Study programme: Title: Information Sciences; specializations in: Archivistics Study, Library Science, Information Sciences (research), Information Sciences (teaching), Museology and heritage management, Informatology; Course: Information Sciences (teaching); Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra informacijskih znanosti (magistar/magistra informacijskih znanosti)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-09-17 15:43:58