| Popis: |
Središte diplomskog rada Emocionalne dimenzije u indeksiranju slika jest emocionalni segment u oznakama slikovnih izvora koji se proučavao i u provedenom istraživanju. Emocije se mogu definirati kao osjećaj koji motivira, organizira i vodi percepciju, misao i akciju (Izard 1991: 14). S druge strane nalazi se pojam iz informacijskih znanosti – indeksiranje. Indeksiranje se odnosi na dodjeljivanje metapodataka entitetima kako bi bilo moguće njihovo pretraživanje (Cleveland i Cleveland 2013: 30). Most koji spaja emocije s indeksiranjem jest afektivno računalstvo koje proučava interakciju ljudi i njihovih emocija s informacijskim sustavima. Istraživanje je obuhvatilo oba područja te su ispitanici trebali dodavati oznake koje se odnose na emocije koje je u njima izazvala odabrana slika. Pred ispitanicima nalazilo se 10 slika kojima su dodali velik broj oznaka. Oznake su analizirane iz više perspektiva: kvantitativne – navedene su oznake prema pojavnosti, kvalitativne – navedene su kvalitativne razlike u izboru emocija za istu sliku, lingvističke – navedene su lingvističke osobitosti poput broja riječi od kojih se sastoji oznaka, padeža, vrste riječi itd., te semantičke perspektive – oznake su pridružene jednoj od semantičkih kategorija (isness, ofness, aboutness te dodatnoj kategoriji emotiveness). The focus of master thesis entitled Emotional dimensions in image indexing is an emotional element of index terms assigned to images, which was analyzed in the research as well. Emotion can be described as a feeling which motivates, organizes, and leads perception, thoughts, and action (Izard 1991: 14). On the other hand, there is a term belonging to the field of information sciences – indexing. Indexing is defined as a process of assigning metadata to entities in order to enable the retrieval of these entities (Cleveland and Cleveland 2013: 86). Connection between emotions and indexing is the field of affective computing. It focuses on the interaction of human beings and their emotions with information systems. This research includes both fields – examinees were asked to assign index terms to ten images in front of them. Index terms should describe the emotions the particular image triggers in them. These index terms were analyzed on multiple levels. Quantitatively, frequency of the terms was analyzed, while on the other hand – qualitatively, the differences in choice of emotion for the same image were analyzed. Linguistically, some linguistic properties such as number of words the index term contains or the usage of case were also analyzed. On semantic perspective, index terms were divided in one of the semantic categories (isness, ofness, aboutness or the additional one – emotiveness). |
