<p>Bioaerosol release in vortex-induced systems, such as those that occur during swirling flows in industrial and domestic settings, is crucial due to its role in airborne microbial dispersion. These emissions are significantly influenced by parameters such as viscosity, temperature, speed, and liquid column height, which modulate shear forces and vortex strength, directly influencing microbial particle release. However, predictive tools to assess the combined effects of these parameters remain limited. This study introduces a framework considering an experimental approach integrated with a machine learning technique to predict bioaerosol release under vortex motion. The performance of ensemble models, including XGBoost and Random Forest, was evaluated in predicting the release of <i>Escherichia coli</i>. The framework uncovers complex nonlinear interactions between operational parameters (viscosity, temperature, rotational speed, and liquid height) and airborne microbial concentrations. XGBoost demonstrated the best predictive performance for bioaerosol emission at various operating parameters, including temperature, viscosity, and rotational speed, with an R² of 0.66 and an RMSE of 3435 CFU/m³, respectively. This framework offers a practical and efficient approach to evaluating bioaerosol emission risks under different conditions. It can be used as a preliminary screening tool during the design of various systems, such as ventilation setups or laboratory spaces, where controlling microbial aerosols is important. The findings support further development of predictive methods in aerosol research for future studies aimed at understanding the physical processes behind bioaerosol generation and dispersion.</p> Graphical Abstract <p></p>