Resumo
Efficiently classifying legal decisions as approved or rejected is critical to ensuring a fair and effective justice system. This study presents a solution for this task, using machine learning techniques. The proposed solution involves data structuring, unit identification,
labeling by experts, and training a machine learning model. O study included an exploratory analysis of source data and pre-processing techniques text for cleaning and normalizing the data. The proposed model achieved a rate high accuracy of 96%. Finally, we validate the model using external data and real cases. The results suggest that the model has the potential to be an effective solution for classifying legal decisions accurately and quickly.