An Unsupervised Method of Exploring Ontologies from Folksonomies

Abstract

Ontology is the basis component of the Semantic Web. Since constructing ontologies is a time consuming job for domain experts, many researches are conducted on automatic extraction of ontologies from texts. As the development of folksonomy, more and more researchers have realized that folksonomy is a better knowledge source for constructing ontologies than texts. Although some works have already been proposed to extract ontologies from folksonomies, these works have little consideration on what a more acceptable and applicable ontology for users should be, and lack an principle to supervise the ontology extraction from a human’s perspective. According to the study in cognitive psychology, there is a family of concepts named basic level concepts, which are frequently used by people in daily life, and most human knowledge is expressed using basic level concepts. In this paper, inspired by studies in cognitive psychology, we try to extract ontologies with basic level concepts from folksonomies. To the best of our knowledge, it is the first work on discovering basic level concepts in folksonomies and using them to construct ontologies.