Using Supervised Classification to Detect Political Tweets with Political Content

• Brunna de Sousa Pereira Amorim UFCG
• André Luiz Firmino Alves IFCE
• Maxwell Guimarães de Oliveira IFCE
• Cláudio de Souza Baptista UFCG

Resumo

Social media platforms have been increasingly used by modern society. In most platforms, users usually share content on various subjects and, in particular, politics is a favorite one. There are many interests in detecting and analyzing such a political content. However, there is a challenge in the process of detecting specific subjects from social media data mainly due to its informality. In this paper, we propose and compare two techniques, based on supervised classification, for the detection of tweets with political content. The results obtained by our approach have demonstrated satisfactory performance, which motivates further research to be undertaken.

Palavras-chave: Machine learning, Deep learning, Logistic regression, Social networks analysis, Text mining