Made available in DSpace on 2015-04-14T14:50:10Z (GMT). No. of bitstreams: 1
457562.pdf: 2425407 bytes, checksum: fefac4edf439614aa48e880ee5b36971 (MD5)
Previous issue date: 2014-01-16 / The task of Relation Extraction from texts is one of the main challenges in the area of Information
Extraction, considering the required linguistic knowledge and the sophistication of the language
processing techniques employed. This task aims at identifying and classifying semantic relations that
occur between entities recognized in a given text. For example, the sentence Next Saturday, Ronaldo
Lemos, director of Creative Commons, will participate in a debate [...]" expresses a institutionalbond"
relation that occurs between the named entities Ronaldo Lemos" and Creative Commons".
This thesis proposes a process for extraction of relation descriptors, which describes the explicit
relations between named entities in the Organization domain (Person, Organization and Location)
by applying, to texts in Portuguese, Conditional Random Fields (CRF), a probabilistic model that
has been used in various tasks e⇥ciently in processing sequential text, including the task of Relation
Extraction. In order to implement the proposed process, a reference corpus for extracting relations,
necessary for learning, was manually annotated based on a reference corpus for named entities
(HAREM). Based on an extensive literature review on the automatic extraction of relations task,
features of different types were defined. An experimental evaluation was performed to evaluate
the learned model utilizing the defined features. Different input feature configurations for CRF were
evaluated. Among them, the highlight was the inclusion of the semantic feature based on the named
entity category, since this feature could express, in a better way, the kind of relationship between the
pair of named entities we want to identify. Finally, the best results correspond to the extraction of
relations between the named entities of Organization and Person categories, in which the F -measure
rates were 57% and 63%, considering the correct and partially correct extractions, respectively. / A tarefa de Extra??o de Rela??es a partir de textos ? um dos principais desafios da ?rea de Extra??o de Informa??o, tendo em vista o conhecimento lingu?stico exigido e a sofistica??o das t?cnicas de processamento da l?ngua empregados. Essa tarefa visa identificar e classificar rela??es sem?nticas que ocorrem entre entidades reconhecidas em um determinado texto. Por exemplo, o trecho No pr?ximo S?bado, Ronaldo Lemos, diretor da Creative Commons, ir? participar de um debate (...)" expressa uma rela??o de v?nculo-institucional" que ocorre entre as entidades nomeadas Ronaldo Lemos" e Creative Commons". Esta tese prop?e um processo para extra??o de descritores de rela??o, os quais descrevem rela??es expl?citas entre entidades nomeadas do dom?nio de Organiza??es (Pessoa, Organiza??o e Local) utilizando o modelo probabil?stico Conditional Random Fields (CRF), e sua aplica??o em textos da L?ngua Portuguesa. O modelo probabil?stico CRF tem sido aplicado eficientemente em diversas tarefas de processamento de texto sequencial, incluindo recentemente a tarefa de Extra??o de Rela??es. A fim de aplicar o processo proposto, um corpus de refer?ncia para extra??o de rela??es, necess?rio para o aprendizado, foi anotado manualmente, tomando como base um corpus de refer?ncia para entidades nomeadas (HAREM). Com base em uma extensa revis?o da literatura sobre a tarefa de extra??o autom?tica de rela??es, features de diferentes naturezas foram definidas. Uma avalia??o experimental foi realizada com o objetivo de avaliar o modelo aprendido utilizando as features definidas. Diferentes configura??es de features de entrada para o CRF foram avaliadas. Dentre elas, destacou-se a inclus?o da feature sem?ntica baseada na categoria da entidade nomeada, j? que essa feature conseguiu expressar melhor o tipo de rela??o que se deseja identificar entre o par de entidades nomeadas. Por fim, os melhores resultados obtidos correspondem ? extra??o de rela??es entre as entidades nomeadas das categorias Organiza??o e Pessoa, na qual as taxas de F-measure foram de 57% e 63%, considerando as extra??es corretas e parcialmente corretas, respectivamente.
| Identifer | oai:union.ndltd.org:IBICT/oai:tede2.pucrs.br:tede/5248 |
| Date | 16 January 2014 |
| Creators | Abreu, Sandra Collovini de |
| Contributors | Vieira, Renata |
| Publisher | Pontif?cia Universidade Cat?lica do Rio Grande do Sul, Programa de P?s-Gradua??o em Ci?ncia da Computa??o, PUCRS, BR, Faculdade de Inform?ca |
| Source Sets | IBICT Brazilian ETDs |
| Language | Portuguese |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, info:eu-repo/semantics/doctoralThesis |
| Format | application/pdf |
| Source | reponame:Biblioteca Digital de Teses e Dissertações da PUC_RS, instname:Pontifícia Universidade Católica do Rio Grande do Sul, instacron:PUC_RS |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | 1974996533081274470, 500, 600, 1946639708616176246 |
Page generated in 0.0019 seconds