Resumo

BLANDON ANDRADE, Juan Carlos  e  ZAPATA JARAMILLO, Carlos Mario. Gate-based Rules for Extracting Attribute Values. Comp. y Sist. [online]. 2021, vol.25, n.4, pp.851-862.  Epub 28-Fev-2022. ISSN 2007-9737.  https://doi.org/10.13053/cys-25-4-3493.

Automated ontology population is intended to enrich ontologies. The main entities of an ontological model are classes, subclasses, attributes (datatype properties), relationships (object properties), and instances. Class instances are important to the scientific community, since some work is devoted to automatically populating ontologies by using statistical methods, information extraction, and natural language processing, among others. The problem is focused on identifying and extracting attribute values of instances. Commonly, such values have a predefined type like numeric, string, boolean, etc. The difficulty arises when you want to know which instance belongs to such values. In this paper we propose an approach based on Natural Language Processing (NLP) and Information Extraction (IE) technologies for extracting attribute values. We use syntactic patterns implemented on the GATE (General Architecture for Text Engineering) tool. The results are independent of the application domain and they exhibit promising values of recall, precision, and F-measure.

Palavras-chave : Automated ontology population; GATE-JAPE patterns; information extraction; natural language processing; ontologies.

        · texto em Inglês     · Inglês ( pdf )