Aluno: Alcides Gonçalves Lopes Junior
Orientadora: Profª. Drª. Mara Abel
Título: A Semantic Neighborhood Approach to Relatedness Evaluation on Well-Founded Domain Ontologies
Linha de Pesquisa: Inteligência Artificial
Local: Prédio 43412 – Sala 215 (sala de videoconferência) – Instituto de Informática UFRGS
Profª. Drª. Viviane Pereira Moreira (UFRGS)
Profª. Drª. Renata Vieira (PUCRS)
Prof. Dr. Marcello Peixoto Bax (UFMG – por videoconferência)
Presidente da Banca: Profª. Drª. Mara Abel
Abstract: In the context of natural language processing and information retrieval, ontologies can improve the results of the word sense disambiguation (WSD) techniques. By making explicit the semantics of the term, ontology-based semantic measures play a crucial role to determine how different ontology classes have a similar or related meaning. In this context, it is common to use semantic similarity as a basis for WSD. However, the measures generally consider only taxonomic relationships, which affects negatively the discrimination of two ontology classes that are related by the other relationship types. On the other hand, semantic relatedness measures consider diverse types of relationships to determine how much two classes on the ontology are related. However, these measures, especially the path-based approaches, has as the main drawback a high computational complexity to calculate the relatedness value. Also, for both types of semantic measures, it is unpractical to store all similarity or relatedness values between all ontology classes in memory, especially for ontologies with a large number of classes. In this work, we propose a novel approach based on semantic neighbors that aim to improve the performance of the knowledge-based measures in relatedness analysis. We also explain how to use this proposal into the path and feature-based measures. Also, we use the ontological meta-property of existential dependence to weight the semantic distance in path-based measures. We evaluate our proposal on WSD using a pre-existent domain ontology for well-core description. This ontology contains 929 classes related to rock facies. Also, we use a set of sentences from four different corpora on Oil&Gas domain. In the experiments, we compare our proposal with state-of-the-art semantic relatedness measures, such as path-based, feature-based, information content, and hybrid methods regarding the F-score, evaluation time, and memory consumption. The experimental results show that the proposed method obtains F-score gains in WSD, as well as a low evaluation time and memory consumption concerning the traditional knowledge-based measures.
Keywords: Knowledge-based measures. Relatedness measures. Semantic neighbors. Ontological meta-properties.