Natural ontologies with elastic matching for elicited knowledge comparison

Issue title: Special section: Selected papers of LKE 2019

Guest editors: David Pinto, Vivek Singh and Fernando Perez

Article type: Research Article

Authors: Calvo, Hiram^*; | Figueroa-Nazuno, Jesús | Mandujano, Ángel

Affiliations: Centro de Investigación en Computación, Instituto Politécnico Nacional, Mexico City, Mexico

Correspondence: [*] Corresponding author. Hiram Calvo, Centro de Investigación en Computación, Instituto Politécnico Nacional, 07738, Mexico City, Mexico. E-mail: [email protected].

Abstract: Natural Ontologies are presented in this work as a useful tool to model the way in which concepts are organized inside the human mind. In order to be compared, ontologies are represented as matrices and an elastic matching technique is used. For this purpose, a distance measure called Modern Fréchet is proposed, which is an approximation to the NP-Complete problem of elastic matching between matrices. An applied case of study is presented in which human knowledge is compared among different groups of people in the Computer Science domain.

Keywords: Natural ontologies, modern fréchet, ontology elicitation, elastic matching, dynamic time warping

DOI: 10.3233/JIFS-179891

Journal: Journal of Intelligent & Fuzzy Systems, vol. 39, no. 2, pp. 2291-2303, 2020