The area of Text Mining has an increasing growing with the support of Machine Learning techniques. Nowadays, the quantity of available data is bigger than the last years. Among the various sources of information, social networks have been highlighted, since they can be used in the extraction of relevant information using posts, tweets and comments as basis. In this work, the social network used is Twitter. Many works explores Twitter platform to get tweets and analyze them to detect social events. This identification is possible due to the users are considered sensors. Consequently, they express their feelings and considerations from external stimuli. The combination of the information and machine learning algorithms makes possible the identification and the monitoring interest events. This work analyzes tweets from Brazil to detect dengue in such country. The results show the utility of the proposal to recognize dengue epidemics in the Brazilian territory.