Publication
Use of discourse knowledge to improve lexicon-based sentiment analysis
| datacite.subject.fos | Humanidades::Outras Humanidades | pt_PT |
| dc.contributor.advisor | Oräsan, Constantin | |
| dc.contributor.advisor | Silva, M ario Jorge Gaspar da | |
| dc.contributor.author | Filho, Pedro Balage | |
| dc.date.accessioned | 2018-04-12T13:20:17Z | |
| dc.date.available | 2018-04-12T13:20:17Z | |
| dc.date.issued | 2012 | |
| dc.date.submitted | 2012 | |
| dc.description | Dissertação de Mestrado, Processamento de Linguagem Natural e Indústria da Língua, Faculdade de Ciências Humanas e Sociais, Universidade do Algarve. School of Law, Social Sciences and Communications, University of Wolverhampton, 2012 | |
| dc.description.abstract | Sentiment Analysis deals with the computational treatment of sentiment in texts. The recent interest for sentiment analysis has grown due the popularity of internet and the increase of user-generated contents, such as blogs, social networks and reviews websites. This work understands sentiment analysis as a classi cation problem. In this problem, a text can be classi ed as positive or negative. Sentiment classi ers can be distinguished by two main approaches: machine learning and lexicon-based. The machine learning approach uses a corpus to automatically learn the best classi cation features. The lexicon-based approach uses a previously computed dictionary with the sentiment lexicon. Discourse is a linguistic level of analysis where the author represents ideas and links concepts in a rational chain of thoughts. One important representation of discourse is the Rhetorical Structure Theory (RST). This theory organizes the discourse in 26 relations that hierarchically represent the text discourse. This objective of this work is to use discourse knowledge to improve a lexicon-based sentiment classi er. To achieve this goal it proposes the SO-RST, a lexicon-based algorithm that weights portions of text under particular RST relations distinctly. Two experiments are reported. The rst experiment veri es if the RST improves sentiment classi cation. It also shows the discourse relations which are most important in the process. The second experiment incorporates discourse markers in the algorithm in order to eliminate the necessity of a RST annotated corpus. It uses the weights learned in the rst experiment to perform the sentiment classi cation. The results obtained showed which RST relations most help the lexicon-based classi er to achieve a better accuracy. The discourse markers introduced in the algorithm showed some directions to follow and the necessary steps to better study this technique. | pt_PT |
| dc.identifier.uri | http://hdl.handle.net/10400.1/10610 | |
| dc.language.iso | eng | pt_PT |
| dc.subject | Análise de sentimentos | pt_PT |
| dc.subject | Análise de sentimentos em léxico | pt_PT |
| dc.subject | Discurso | pt_PT |
| dc.subject | Teoria da estrutura retórica | pt_PT |
| dc.title | Use of discourse knowledge to improve lexicon-based sentiment analysis | pt_PT |
| dc.type | master thesis | |
| dspace.entity.type | Publication | |
| rcaap.rights | restrictedAccess | pt_PT |
| rcaap.type | masterThesis | pt_PT |
| thesis.degree.discipline | Processamento de Linguagem Natural e Indústria da Língua | |
| thesis.degree.grantor | Universidade do Algarve. Faculdade de Ciências Humanas e Sociais | |
| thesis.degree.level | Mestre | |
| thesis.degree.name | Mestrado em Processamento de Linguagem Natural e Indústria da Língua | pt_PT |