Análise de Sentimento em Redes Sociais para a Língua Portuguesa Utilizando Algoritmos de Classificação

  • Erikson Júlio de Aguiar
  • Bruno S. Faiçal
  • Jó Ueyama
  • Glauco Carlos Silva
  • André Menolli

Resumo


Sentiment Analysis in social network has been explored in different types of research, and its main intention is to extract user's opinion on a wide range of subjects, making possible to obtain significant information. For Portuguese, Sentiment Analysis research is still being established. In this context, this work proposes a method to estimate sentiment in social networks for the Portuguese language, focusing on Twitter. For such purpose, we used a Committee, which is implemented through a set of machine learning algorithms for classification. The evaluation of the proposed method was performed using statistical and performance tests. The results indicate that the Committee had better accuracy when compared to other machine learning algorithms for the performance tests. Nevertheless, there was no proven statistical difference between the Committee's algorithm and some of the algorithms, indicating that these methods could achieve equivalent accuracy to the Committee in some specific situations, such as a larger database.
Publicado
10/05/2018
Como Citar

Selecione um Formato
AGUIAR, Erikson Júlio de; FAIÇAL, Bruno S.; UEYAMA, Jó; SILVA, Glauco Carlos; MENOLLI, André. Análise de Sentimento em Redes Sociais para a Língua Portuguesa Utilizando Algoritmos de Classificação. In: SIMPÓSIO BRASILEIRO DE REDES DE COMPUTADORES E SISTEMAS DISTRIBUÍDOS (SBRC) , 2018 Anais do XXXVI Simpósio Brasileiro de Redes de Computadores e Sistemas Distribuídos. Porto Alegre: Sociedade Brasileira de Computação, may 2018 . ISSN 2177-9384.