A System for Specifying Semantic Relations for Knowledge Representation

Abstract

Semantic relations are fundamental for understanding the nature of the connection between two concepts in a domain. This paper presents a model for extracting semantic relations for the representation of knowledge

from academic documents in the context of the Portuguese language. A Web information system called Se- mantizar was developed to support the extraction of semantic relations from classificatory structures that rep- resent specific academic documents. To evaluate the qualitative performance of Semantizar, a case study was

carried out, which pointed to important contributions to research about semantic relations extraction. Accord- ing to outcomes, when two concepts of a classificatory structure exist in a sentence, a semantic relationship

between them can actually exist. Finally, it is concluded that this research is relevant because it brings im- portant findings for the extraction of semantic relations for the knowledge representation of academic docu- ments to be applied in the Brazilian scenario.