Prediction of pharmaceuticals occurrence based on sales data and Machine learning algorithms.

<b>anvisa2014_2020</b>: This is a pre processed data from the sales data of industrial pharmaceuticals from SNGPC, Brazil's data.<b>preprocessing_anvisa_ana_datasets</b>: the data manipulation of ANVISA and ANA data to the final dataset anvisa2014_2020. <b>Antibioticos/Carbamazepina</b>: contains the main codes of the prediction models to classify the occurrence concentrations of some antibiotics and Carbamazepine, by tree boosting algorithms.<b>Sensitivity Analysis</b><b>cargas25_50_75_AS:</b> contains the pre processed updated dataset of pharmaceuticals sales data for the 3 scenarium conditions.<b>carb_AS/ antibioticos_AS</b>: the final integrated dataset used for modeling Carbamazepine and Antibiotics sensitivity models. These datasets can be executed on the same code of the base models.