Sentiment Analysis of Twitter Posts About the 2017 Academy Awards
Resumo
This paper aims to perform the sentiment analysis of Twitter posts related to the movies nominated for Best Picture of the 2017 Oscars in order to find out if there is a correlation between the posts and the Oscar winners. A tweets database was built, pre-processed, and later evaluated by three distinct approaches: Naive Bayes, Distant Supervision Learning, and Polarity Function. It was possible to predict which movie would be considered the winner and which would be among the less prestigious ones. It was noted that Twitter users prefer to post positive comments about movies rather than saying bad things about the ones they did not like. Furthermore, it was verified that award shows such as the Oscars cause a growth in the number of posts on Twitter.
Referências
Almeida, R. J. d. A. (2012). Estudo da ocorrência de cyberbullying contra professores na rede social Twitter por meio de um algoritmo de classificação Bayesiano. Texto Livre: Linguagem e Tecnologia, 5(1):1–7.
Baeza-Yates, R. and Ribeiro-Neto, B. (2013). Recuperação de Informação: Conceitos e Tecnologia das Máquinas de Busca. Bookman Editora, Porto Alegre, RS, Brazil, 2 edition.
Benevenuto, F., Almeida, J. M., and Silva, A. S. (2011). Explorando Redes Sociais Online: Da Coleta e Análise de Grandes Bases de Dados às Aplicações. In Livro Texto de Minicursos - SBRC 2011, chapter 2, pages 63–101. Campo Grande, MT, Brazil.
Bothos, E., Apostolou, D., and Mentzas, G. (2010). Using Social Media to Predict Future Events with Agent-Based Markets. IEEE Intelligent Systems, 25(6):50–58.
Cetinsoy, A. (2017). Predicting the 2017 Oscar Winners with BigML. https://dzone.com/articles/predicting-the-2017-oscar-winners. Accessed on 26 Feb. 2017.
De Smedt, T. and Daelemans, W. (2012). VewVreselijk mooi! (terribly beautiful): A Subjectivity Lexicon for Dutch Adjectives. Proceedings of the 8th International Conference on Language Resources and Evaluation (LREC’12), pages 3568–3572.
DeepDive (2017). Distant Supervision. http://stanford.io/2Br4o1E. Accessed on 01 Nov. 2017.
Felix, N. (2016). Análise de sentimentos em textos curtos provenientes de redes sociais. PhD thesis, Universidade de S˜ao Paulo - S˜ao Carlos, S˜ao Paulo, SP, Brazil.
Go, A., Bhayani, R., and Huang, L. (2017). A Twitter Sentiment Analysis Tool. http://help.sentiment140.com/home. Accessed on 18 Apr. 2017.
Go, A., Huang, L., and Bhayani, R. (2009). Twitter Sentiment Classification using Distant Supervision. Processing, 150(12):1–6.
Guilford, J. P. (1957). Fundamental statistics in psychology and education. Science Education, 41(3):565.
Krauss, J., Nann, S., and Simon, D. (2008). Predicting movie success and academy awards through sentiment and social network analysis. In 16th European Conference on Information Systems, page 12.
Loria, S., Keen, P., Honnibal, M., Yankovsky, R., Karesh, D., Dampsey, E., Childs, W., Schnurr, J., Qalieh, A., Ragnarsson, L., Coe, J., Calvo, A., Kulshrestha, N., Eslava, J., and Albert, J. (2017). TextBlob: Simplified Text Processing. http://textblob.readthedocs.io/en/dev/index.html. Accessed on 28 Aug. 2017.
Pak, A. and Paroubek, P. (2010). Twitter as a Corpus for Sentiment Analysis and Opinion Mining. In Proceedings of the 7th Conference on International Language Resources and Evaluation, pages 1320–1326.
Ribeiro, L. B. (2015). Análise de sentimento em comentários sobre aplicativos para dispositivos móveis. Undergraduate paper. Universidade de Brasília, Brasília, DF, Brazil.
Schmitt, V. F. (2013). Uma Análise Comparativa De Técnicas De Aprendizagem De Máquina Para Prever a Popularidade De Postagens No Facebook. Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil.
Teixeira, D. and Azevedo, I. (2011). Análise de opini˜oes expressas nas redes sociais. RISTI - Revista Ibérica de Sistemas e Tecnologias de Informação, 8:53–65.
Wong, V. (2013). How Oscar Nominations Affect the Box Office. https://www.bloomberg.com/news/articles/2013-01-10/how-oscar-nominationsaffect-the-box-office. Accessed on 22 Apr. 2017.