Sentiment Analysis in Tweets: a Case Study on Federal Institutions Bugdet Cuts

• Danielly Rayanne M. Lima IFPB
• Raissa Ohana F. O e Silva IFPB
• Álvaro Getúlio Medeiros IFPB
• André Luiz Alves IFPB

DOI: https://doi.org/10.5753/encompif.2020.11064

Abstract

Understanding the opinions of citizens, especially on the part of governments, is fundamental to decision-making. However, with the large and massive volume of data, process this information manually is a complicated task and does not provide satisfactory results. This research work to build a sentiment classifier enables analyzing opinions automatically using tweets with context of provision cuts made by the brazilian government in the half year of 2019. In order to solve the problem, four machine learning techniques for natural language processing were formulated, where the technique that proved the best result presented an accuracy of 72%.

Keywords: Sentiment analysis, Machine Learning, Natural Language Processing

References

EIRINAKI, Magdalini, PISAL, Shamit, SINGH, Japinder (2012). Feature-based opinion mining and ranking. Journal of Computer and System Sciences, Vol 78, p. 1175–1184.

LIU, N., & Shen, B. (2020). ReMemNN: A novel memory neural network for powerful interaction in aspect-based sentiment analysis. Neurocomputing.

O'REILLY, Tim (2007). What Is Web 2.0: Design Patterns and Business Models for the Next Generation of Software. Communications & Strategies, Vol 1, p. 17-37.

SILVA, N. F. F. Um estudo comparativo sobre métodos de análise de sentimentos em tweets. Revista de Sistemas de Informação da FSMA, v. 1, n. 22, p. 35-48, 2018.

AGUIAR, Erikson Júlio et al. Análise de sentimento em redes sociais para a língua portuguesa utilizando algoritmos de classificação. In: Anais do XXXVI Simpósio Brasileiro de Redes de Computadores e Sistemas Distribuídos. SBC, 2018.