Naslov
Ontological model of the emotion lexical concepts and constructions: from data to knowledge based visualization.

Autori
Perak, Benedikt ; Damčević, Katarina

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, ostalo

Izvornik
Visual Digital Humanities: Representing and Interpreting Humanities Data / - Tartu, 2016

Skup
Visual Digital Humanities: Representing and Interpreting Humanities Data

Mjesto i datum
Tartu, Estonija, 12.10.2016. - 14.10.2016

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
digital humanities, visualization

Sažetak
Language is a symbolic tool that enables human representation and categorization of the world (Langacker 2008). This implies that language conveys knowledge about the world, but also models our representation of the world (Casasanto 2016) via conventionalized linguistic expressions. According to the cognitive linguistic approach of emotion expression (Perak 2016), the conceptualization of emotional content using emotional lexemes and syntactic-semantic constructions can be analysed in terms of meronymic (metonymic part- whole) profiles of the emotion event (Scherer 2009) and metaphoric mappings (Indurkhya 1992) that violate the ontological relations between the constituents of the construction. The aim of this presentation is to visualize the results of a) data mining, b) construction collocation analysis and c) ontological classification procedures of English (enTenTen 13 Gw) and Croatian (hrWac 1, 2 Gw) corpora, representing distinctive syntactic-semantic- pragmatic properties of 10 lexical concepts of emotion. Every stage of the research is represented with appropriate sets of visualization tools: a) statistical histograms, b) force based graph layouts and c) ontology based graphs. These different types of visualizations reveal the shift from quantitative to mixed quantitative-qualitative representation of data, which could be termed: from information about the linguistic use towards the representation of linguistic meaning. We will demonstrate the distinction between these mixed qualitative and quantitative methods of describing knowledge and standard linguistic usage obtained by tools for data mining (Sketch-engine), ontological classification of domains (Protégé), and visualization of analysis and classification (Gephi, D3 based interactive visualization scripts) arguing for the need to introduce an ontological model of lexical concepts and constructions (OMLCC). OMLCC enables cross- linguistic formalization of the creative semasiological processes (etymology, metonymy, metaphor extensions) that language(s) as an instrument of cognizing and cultural conceptualization tool adds to the representation of emotional knowledge.

Izvorni jezik
Engleski

Znanstvena područja
Informacijske i komunikacijske znanosti, Filologija

POVEZANOST RADA