Application of machine learning for the prediction of Bacillus cereus growth: A study on the integration of mathematical models.

In this study, machine learning was used to predict Bacillus cereus growth trends across various food matrices using data from the ComBase database. The developed framework features a predictive model for log growth values and a classification model for risk assessment developed using the Neural Designer and Colaboratory (Colab) platforms. The data quality was improved through k-means clustering and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) for filtering, whereas dimensionality reduction via principal component analysis, followed by DBSCAN facilitated better outlier detection. To effectively capture microbial growth dynamics, mathematical models such as the Gompertz and Double Weibull models were integrated into the machine learning pipeline. Experimental validation confirmed the practical utility of the model and demonstrated its ability to predict B. cereus growth under diverse conditions.