Infections by Staphylococcus aureus are a serious healthcare problem, with a high alert for resistant strains. The World Health Organization characterized the methicillin-resistant S. aureus in the high priority group for the development of new antibiotics. Following this need, inhibition of DNA gyrase presents itself as an interesting drug target, due to the lack of homologs in mammalian, and it could be a way to overcome the resistance problem. In this study, classification structure-activity relationship models based on the Random Forest algorithm were employed to classify antibacterial compounds acting as DNA gyrase inhibitors. The models were generated and validated for the classification of antibacterial activity (external MCC = 0.775), DNA gyrase inhibition (external MCC = 0.577), and a consensus of these two endpoints (external MCC = 0.577). The structural interpretation highlighted the relevance of heterocycles substituents. This information may provide understanding in the structure-activity relationship of this compounds class, providing insights for further developments.