Rockbursts are known for their unpredictable and violent nature, representing a significant threat to workers’ safety, mining productivity, and operational costs. Therefore, a quantitative assessment of rockburst damage is significant for geotechnical risk management in seismically active underground mines. Over the past few decades, numerous studies have been conducted to mitigate the risk posed by rockburst from various perspectives. Despite the scientific achievements and technological advances in ground control, rockburst still threatening underground mine operations because of the elusive character of the rockburst phenomenon and the challenges associated with its reliable prediction. Hence, the current study examines the possibility of implementing supervised machine learning algorithms to classify seismic events. Mining-induced seismicity pertaining to a deep gold mine exploiting the Witwatersrand Basin of South Africa was used to implement the models. The validation results showed the classification accuracy varied between 70 and 84% depending on the model implemented. These indicate good agreement with the seismic data and the induced rockburst events. It was concluded that the results of the study could assist in minimizing the risk of rockbursting in deep mines.