Detalhes bibliográficos
| Ano de defesa: |
2015 |
| Autor(a) principal: |
Freitas, Larissa Astrogildo de
 |
| Orientador(a): |
Vieira, Renata |
| Banca de defesa: |
Não Informado pela instituição |
| Tipo de documento: |
Tese
|
| Tipo de acesso: |
Acesso aberto |
| Idioma: |
por |
| Instituição de defesa: |
Pontifícia Universidade Católica do Rio Grande do Sul
|
| Programa de Pós-Graduação: |
Programa de Pós-Graduação em Ciência da Computação
|
| Departamento: |
Faculdade de Informática
|
| País: |
Brasil
|
| Palavras-chave em Português: |
|
| Área do conhecimento CNPq: |
|
| Link de acesso: |
http://tede2.pucrs.br/tede2/handle/tede/6031
|
Resumo: |
Sentiment Analysis is the field of study that analyzes people’s opinions in texts. In the last decade, humans have come to share their opinions in social media on the Web (e.g., forum discussions and posts in social network sites). Opinions are important because whenever we need to take a decision, we want to know others’ points of view. The interest of industry and academia in this field of study is partly due to its potential applications, such as: marketing, public relations and political campaign. Research in this field often considers English data, while data from other languages are less explored. It is possible realize data analysis in different levels, in this work we choose a finer-grain analysis, at aspect-level. Ontologies can represent aspects, that are “part-of” an object or property of “part-of” an object, we proposed a method for feature-level sentiment analysis using ontologies applied to Brazilian Portuguese reviews. In order to obtain a complete analysis, we recognized features explicit and implicit using ontologies. Relatively less work has been done about implicit feature identification. Finally, determine whether the sentiment in relation to the aspects is positive or negative using sentiment lexicons and linguistic rules. Our method is comprised of four steps: preprocessing, feature identification, polarity identification and summarizing. For evaluate this work, we apply our proposal method to a dataset of accommodation sector. According to our experiments, in general the best results were obtained when using TreeTagger, synsets with polarities from Onto.PT and linguistic rule (adjective position) for negative polarity identification and (baseline) for positive polarity identificatio |
