Scientific articles published in the period from 1996 to 2018 are evaluated in order to identify the incidence records of bovine tuberculosis and its relationship with some factors related to its behavior. The research was carried out using text mining techniques. The trends were estimated using the linear regression and generalized additive models. All the analyzes were performed in the statistical computing environment R using the packages tm, ggplot2, factoextra, factominer and string. The results validate the hypothesis of the possibility of identifying bTB incidence records using published articles. In addition, the interrelation of the evaluated variables is established in order to allow the identification of the efforts made and the results achieved, and, equally, to deduce the future behavior from disease registers, which favors the establishment of corrective measures in the established policies.