Visual mapping art : data visualization & crowdsourcing for an ostensive definition of art

• Main Author: Ong, Zi Feng
• Other Authors: Bernhard Johannes Schmitt
• Format: Thesis-Master by Research
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Visual arts and music::Arts in general; Library and information science::Libraries::Cataloguing and classification
• Online Access: https://hdl.handle.net/10356/168961
• Institution: Nanyang Technological University

Art categorization is challenging (Novitz, 1996; Ziff, 1953) because definitions of art vary throughout art history and amongst different philosophical schools. (Adajian, 2018, Tatarkiewicz, 1980). The purpose of this research is not to join these art historical and philosophical discussions, but to build a perceptible, practicable component that could be used by the public through applying the ostensible definitions of art as an exploratory, statistical database. This research will build the framework to categorize forms of art in a database as the practical component. The core data will be gathered from existing art databases, such as the Library of Congress (LCC) (Class-N - Fine Arts) and the Universal Decimal Classification (UDC) (73/76 Various arts & crafts); and controlled vocabulary systems such as Art & Architecture Thesaurus (AAT). The core data will be utilized as a framework for crowdsourcing, with Human Intelligence Tasks (HIT) implemented on Amazon Mechanical Turk (MTurk) to increase the number of art forms in the database to include less prevalent art forms. As an endpoint of this research, the data obtained through the process of existing art database extraction and crowdsourcing will be shown through a network visualization on a website as a practical component for the public to observe the classification of art in a graphical fashion.