Extracción y organización del conocimiento de etiquetadosaplicación a etiquetados en repositorios digitales sobre arte

Abstract

Formal Concept Analysis (FCA) is a branch of Applied Mathematics whose aim is to discover, extract and organize knowledge from data. FCA provides techniques and tools to reason with such knowledge. The aim of this paper is to analyse the applicability of FCA to cultural digital repositories that use tagging. It is possible to study the structure of concepts implicit in the tags by means of FCA, in such way that it is possible to detect some patterns allowing the estimation of the soundness of these tag set. Among other examples, we apply results from (Aranda Corral et al., 2012a) to analyse two repositories about art: Baroque Art from CulturePlex Lab (http://baroqueart.cultureplex.ca/) and the Visual Archive of Gothic Architecture and Sculpture in Ireland (http://www.gothicpast.com/). The conceptual structure extracted from both repositories is compared with the one from WordNet (http://wordnet.princeton.edu/), which is an example of successful semantic representation. Under the called Scale-Free Conceptualization hypothesis (SFCH) (Aranda Corral et al., 2012a), the soundness of tagging sets (folksonomies) is estimated.