Analysis of Feeling of Insecurity Through Twitter
Abstract
The sense of insecurity has a direct influence on the quality of life of the citizen. This work used machine learning algorithms to identify the feeling of insecurity of the Brazilian citizen, express posts on the social network Twitter. For this, a balanced database was generated with 400 tweets, which were previously classified between feelings of insecurity or others. Prediction models were created to classify the tweets posted in the network with the algorithms FlorestaAleatória, SVM and Logistic Regression. The algorithms reached indexes of up to 0.34 for the Matthews correlation coefficient and 0.69 for the F1 measure. A web application was developed for online ranking of the new tweets published using a better three strategy, demonstrating the viability of using such an approach to accompany the sense of insecurity among the users of this network.
References
social. Departamento de Estudos da Escola Superior de Guerra. Rio de Janeiro, 2011.
Disponível em:
em: 11 out. 2017.
Aquino, P., & Batista, V. (2017). Sistema de minería de opiniones para el análisis de
sentimiento en Twitter. Avances en Informática y Automática. Décimo Workshop,
pp.38-54. ISBN 978-84-617-9707-3. Retrieved from http://hdl.handle.net/10366/133039
Bicalho, J. C. R. e Gastalho, P. P. Insegurança e produção de subjetividade no Brasil
contemporâneo. Universidade Federal do Rio de Janeiro. Revista de Psicologia, v. 21
– n. 2, p. 425-444. Instituto de Psicologia, 2009. Disponível em:
Bishop, C. M. Pattern Recognition and Machine Learning. Springer. 2006
Bo Pang and Lillian Lee (2008), Opinion Mining and Sentiment Analysis, Foundations
and Trends® in Information Retrieval: Vol. 2: No. 1–2, pp 1-135. http://dx.doi.org/10.1561/1500000011
Caragea, C. et al Mapping Moods: Geo-Mapped Sentiment Analysis During Hurricane
Sandy. In: 11th International Conference on Information Systems for Crisis Response and Management. 2014. Conference Proceedings.
Frattari, N. F. Insegurança: as práticas e discursos do medo na cidade de Goiânia. Universidade Federal De Goiás. Goiânia, 2009. Disponível em:
set. 2017.
Joachims, T. Text Categorization with Support Vector Machines: Learning with Many
Relevant Features. Proceedings of ECML, Springer, 1998. Disponível em:
29 set. 2017.
Madeira, R. O. C. Aplicação de técnicas de mineração de texto na detecção de
discrepâncias em documentos fiscais do Rio de Janeiro. Escola de Matemática
Aplicada da Fundação Getúlio Vargas. Rio de Janeiro, 2015. Disponível em:
Maia, P. et al.Application of text mining techniques for classification of documents: a
study of automation of complaints screening in a Brazilian Federal Agency. Universidade de Brasília. Departamento de Ciência da Computação, 2013.Disponível
em: http://bdm.unb.br/bitstream/10483/7711/1/2013_fernandoleandrodossantos.pdf>.
Acesso em: 25 out. 2017.
Matthews, B. W.. Comparison of the predicted and observed secondary structure of T4
phage lysozyme. Biochimica et Biophysica Acta (BBA) - Protein Structure. 1975.
Mitchell, T. Machine Learning. McGraw Hill. 1997
Porotsky, S. Bluvband, Z. Advanced Text Mining Algorithms for Aerospace Anomaly
Identification.Proceedings of ESREL Conference. 2011. DOI: 10.1201/b11433-54.
Acesso em: 08 out. 2017.
Powers, D. Evaluation: From Precision, Recall and F-Factor to ROC, Informedness,
Markedness & Correlation. Technical Report SIE-07-001. 2007. Disponível em:
Santos, F.. Mineração de opinião em textos opinativos utilizando algoritmos de
classificação. Universidade de Brasília. Departamento de Ciência da Computação,
2013. Disponível em:
Acesso em: 08 out. 2017.
Sena, J. A., Lopes, K. T. e Pereira, A. O transtorno de estresse pós-traumático e a
violência urbana. Psicologia ciências iológicas e da saúde. Maceió, 2013. Disponível
em:
em: 25 set. 2017.
Silva, H. C. Políticas públicas para o idoso: marcos referenciais no trato da violência no
município de Florianópolis. Universidade Federal de Santa Catarina, 2007.
Disponível em:
Terpstra, T. et al. Towards a realtime Twitter analysis during crises for operational crisis
management. In: 9 th International Conference on Information Systems for Crisis Response and Management. 2012. Conference Proceedings.
Zielinski, A. et al. Social Media Text Mining and Network Analysis for Decision
Support in Natural Crisis Management. In: 10th International Conference on Information Systems for Crisis Response and Management. 2012. Conference Proceedings.
