Računalno potpomognuta analiza karakterizacije mitskih likova u romanima Neil Gaimana

master's thesis

• Title (english): Computer-aided analysis of mythical figures characterisation in Neil Gaiman's novels
• Author: Mila Bikić
• Mentor: Kristina Kocijan (mentor)
• Committee member: Vedran Juričić (predsjednik povjerenstva)
• Committee member: Vjera Lopina (član povjerenstva)
• Committee member: Kristina Kocijan (član povjerenstva)
• Granter: University of Zagreb; Faculty of Humanities and Social Sciences; (Department of information and Communication sciences); Zagreb
• Defense date and country: 2024-09-18, Croatia
• Scientific / art field, discipline and subdiscipline: SOCIAL SCIENCES; Information and Communication Sciences; Organization and Informatics

Abstract

Obrada prirodnog jezika je područje koje kombinira računalnu lingvistiku, umjetnu inteligenciju i strojno učenje kako bi omogućila računalima analizu i interpretaciju ljudskog jezika. Ono uključuje razne primjene kao što su pretraživanje informacija, strojno prevođenje i ekstrakcija informacija, ali može se primijeniti i na humanistička istraživanja. U ovom radu fokus je na analizi romana Neil Gaimana Dobri predznaci i Američki bogovi korištenjem metoda iz računalne i korpusne lingvistike. Cilj istraživanja je bio ispitati kako se pridjevi koriste za opisivanje likova i kako to odražava šire teme i poruke unutar tih djela. Istraživanje je podijeljeno u tri faze. U prvoj fazi prikupljen je i očišćen korpus, koristeći programe Calibre i Notepad++. Druga faza uključivala je izradu rječnika te sintaktičkih gramatika za prepoznavanje pridjeva i likova na koje se odnose u programu NooJ. Treća faza obuhvaćala je rješavanje problema koreferencije primjenom strojnog učenja u programskom jeziku Python. Rezultati upućuju na razlike u načinu opisivanja likova između dva romana. U Američkim bogovima dominira detaljna karakterizacija glavnih likova kroz pridjeve, dok se u Dobrim predznacima koristi humorističan stil pisanja koji ne oslanja se na opis likova kroz pridjeve. Ovi uvidi ukazuju na različite pristupe u književnoj analizi karakterizacije likova, od detaljnih opisa u Američkim bogovima do kreativnih stilskih figura u Dobrim predznacima.

Abstract (english)

Natural Language Processing is a discipline that combines computational linguistics, artificial intelligence, and machine learning to enable computers to analyse and interpret human language. It encompasses various applications such as information retrieval, machine translation, and information extraction, and can also be applied to humanities research. This paper focuses on the analysis of Neil Gaiman's novels Good Omens and American Gods using methods derived from computational and corpus linguistics. The goal of the research was to examine how adjectives are used to describe characters and how this reflects broader themes and messages within these works. The research was divided into three phases. In the first phase, a corpus was collected and cleaned using Calibre and Notepad++ programs. The second phase involved creating dictionaries and syntactic grammars to identify adjectives and the characters they describe in the NooJ program. The third phase addressed the problem of coreference resolution through the application of machine learning in the Python programming language. The results indicate differences in the way characters are described between the two novels. In American Gods, there is a dominance of detailed characterization of the main characters through adjectives, while in Good Omens, a humorous writing style is used that does not rely on describing characters through adjectives. These insights point to different approaches in literary analysis of characterization, from detailed descriptions in American Gods to creative stylistic figures in Good Omens.

• Keywords: digitalna humanistika; Dobri predznaci; Američki bogovi; Neil Gaiman; računalna lingvistika; NooJ; Python
• Keywords (english): digital humanities; Good Omens; American Gods; Neil Gaiman; computational linguistics; NooJ; Python
• Language: croatian
• URN:NBN: urn:nbn:hr:131:641132
• 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 (research); 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: 2024-09-18 11:35:44