Evaluation of methods for taxonomic relation extraction from text

• Ano de defesa: 2015
• Autor(a) principal: Granada, Roger Leitzke
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Pontifícia Universidade Católica do Rio Grande do Sul; Faculdade de Informática; Brasil; PUCRS; Programa de Pós-Graduação em Ciência da Computação
• Palavras-chave em Português: ONTOLOGIA; PROCESSAMENTO DA LINGUAGEM NATURAL; INFORMÁTICA; CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://tede2.pucrs.br/tede2/handle/tede/7108

Modern information systems are changing the idea of “data processing” to the idea of “concept processing”, meaning that instead of processing words, such systems process semantic concepts which carry meaning and share contexts with other concepts. Ontology is commonly used as a structure that captures the knowledge about a certain area via providing concepts and relations between them. Traditionally, concept hierarchies have been built manually by knowledge engineers or domain experts. However, the manual construction of a concept hierarchy suffers from several limitations such as its coverage and the enormous costs of extension and maintenance. Furthermore, keeping up with a hand-crafted concept hierarchy along with the evolution of domain knowledge is an overwhelming task, being necessary to build concept hierarchies automatically. The (semi-)automatic support in ontology development is usually referred to as ontology learning. The ontology learning from texts is usually divided in steps, going from concepts identification, passing through hierarchy and non-hierarchy relations detection and, seldom, axiom extraction. It is reasonable to say that among these steps the current frontier is in the establishment of concept hierarchies, since this is the backbone of ontologies and, therefore, a good concept hierarchy is already a valuable resource for many ontology applications. A concept hierarchy is represented with a tree-structured form with specialization/generalization relations between concepts, in which lower-level concepts are more specific while higher-level are more general. The automatic construction of concept hierarchies from texts is a complex task and since the 1980 decade a large number of works have been proposing approaches to better extract relations between concepts. These different proposals have never been contrasted against each other on the same set of data and across different languages. Such comparison is important to see whether they are complementary or incremental, also we can see whether they present different tendencies towards recall and precision, i.e., some can be very precise but with very low recall and others can achieve better recall but low precision. Another aspect concerns to the variation of results for different languages. This thesis evaluates these different methods on the basis of hierarchy metrics such as density and depth, and evaluation metrics such as Recall and Precision. The evaluation is performed over the same corpora, which consist of English and Portuguese parallel and comparable texts. Both automatic and manual evaluations are presented. The output of seven methods are evaluated automatically and the output of four methods are evaluated manually. Results shed light over the comprehensive set of methods that are the state of the art according to the literature in the area.