Food safety and risk assessment is vital to the protection of public health. Large amounts of food safety-related data are generated daily from facility inspections, analytical testing, disease surveillance, and even social media. Advanced data analytic techniques, including machine learning, deep learning, and natural language-processing algorithms, helps in efficiently analyzing complex data and can be applied to these datasets for food safety prediction. This review provides an overview of how whole genome sequencing, product inspection, disease surveillance, infrared spectra, and hyperspectral imaging data are being used to develop machine learning and deep learning models for the prediction of potential hazards, risk assessments, and food source attribution. All rights reserved, Elsevier.

You do not currently have access to this content.