Extracting automatically a domain ontology from the “Book of Properties” of the archbishop’s table of Braga

Abstract

During the last years, many researchers used ontology-based technologies for developing ontology-learning processes for acquiring knowledge included in textual information. In our case, we focused such processes to study the contents of Portuguese manuscripts from the 17th century, aiming to explore the formalities of the language and characterize the elements contained in the texts. In particular, we worked over a very specific manuscript. In this paper, we present and describe the implementation of a semi-automatic ontology learning process for extracting the knowledge contained in the “Book of Properties”. This manuscript includes a detailed inventory of assets belonging to the Archbishop’s Table of Braga at the beginning of the 17th century. Our goal is to provide a fundamental computational instrument for studying and learning the geography, culture, agriculture, history, and genealogy in Portugal at that time.