Intelligent System to Support Critical Analysis of Parliamentary Body Performance
Abstract
In this work, we propose a method for analyzing and processing textual news from the Legislative Assembly of Sergipe State website, by means of a text mining process which can be applied to legislative chambers from different states. The main goal is to facilitate to the voter the documentation of such news, and provide decision support on which candidate to vote for state representative. The analysis consists mainly in the automatic categorization of news on predetermined labels that, aligned with other data, measure the performance of parliament during its mandate by generating an index, the QoP (Quality of Parlamentarian). Different techniques have been used to categorization and Multinomial Naive Bayes algorithm with a minimum cutoff frequency equal to 0 (zero) has obtained the highest success rate, in the order of 84%. The QoP index was effective for the dynamic ranking of deputies according to the assessment of their quality in both quantitative and qualitative terms. A Web application was developed and allows the voters to view the default ranking as well as generate alternative rankings through a custom interface where they can modify weights used to calculate the index
References
KONCHADY, Manu. Text Mining Application Programming. Boston: Ed. Thomson, 2006. 412 p. 95
LEHAL, Gurppreet S. GUPTA, Vishal. A Survey of Text Mining Techniques and Applications. Journal of Emerging Technologies in Web Intelligence, Vol 1, No 1, 60-76, 2009.
PAK, Albert Frederico de Menezes Il. Aplicação de Técnicas de Mineração de Texto para Categorização de Eventos de Segurança no CTIR Gov. 2010. Xi, 71 f. Dissertação (Mestrado em Informática) – Universidade Federal de Brasília, Brasília, 2010.
RIZZI, Claudia Brandelero (2000). et al. Fazendo uso da categorização de textos em atividades empresariais. Disponível em: <http://leandro.wives.nom.br/ptbr/publicacoes/iskmdm2000-2.pdf>. Acesso em: 29 mar. 2011.
OLIVEIRA, C. J. S. & ARAÚJO, A. de A. Classificando Imagens Coletadas na World Wide Web em duas Classes Semânticas: Imagens Gráficas e Imagens Fotográficas, Relatório Técnico RT.DCC.006/2002, DCC/ICEx/UFMG, Belo Horizonte, MG, Brasil, Dezembro, 2002, 29p.
STEINBRUCH, David. Um estudo de Algoritmos para Classificação Automática de Textos Utilizando naive-Bayes. Pontifícia Universidade Católica, Rio de Janeiro, RJ, 2006.
TOMAZELA, Maria das Graças J. M.; DANIEL, Luiz Antônio. Uma Estratégia de Preparação de Dados para Aumento de Precisão de Modelos de Classificação da Produtividade de Cana-de-açucar. Faculdade de Tecnologia de Indaiatuba, Indaiatuba, SP, 2011.
WITTEN, Ian H.; FRANK, Eibe; HALL, Mark A. Data Mining: Practical Machine Learning Tools and Techniques. 3rd edition. Hamilton, New Zealand: Ed. Morgan 2011. Kaupmann,
