Risk assessement for food safety and public health: a machine learning approach

• Autor(a) principal: Silva, Maria Clara Ferreira e
• Data de Publicação: 2022
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.22/21450
• topic: Food safety; Public health; ASAE; Risk assessement; Machine learning

Foodborne diseases continue to spread widely in the 21st century. In Portugsl. The economic and Food Safety Authority (ASAE), have the goal to monitoring and preventing non-compliance with regulatory legislation and food safety, regulate the conduct of economic activities in the food and non-food sectors, as well as acess and comunicate risks in the food chain. In this work, it was evaluated the global risk considering three risk factors provided by ASAE (non-compliance rate, producto or service risk and consumption volume). It was also compared the performance on the prediction of risk of four classification models Decision Tree, Naive Bayes, K-Nearest Neighbor and Artificial Neural Network before and after feature selection and hyperparameter tuning. Our principal findings revealed that the servisse Provider food and beverage and retail were the activity sectors presente in the dataset with the highest global risk associated with. It was also observed that the Artificial Neural Network classifier presented the best results of 60%, however it was the model that took longer to train. It was also detected that the Chi-square feature selection method provided better results than the ANOVA F-test. It was also verified that data balancing using the SMOTE method led to a performance increase of 90% with the Decision tree and K-Nearest Neighbor modelas. This work allowed to conclude that the use of machine learning can be helpful in risk assessment related to food safety and public health. It was also concluded that the áreas regarding major global risks are the ones which are more frequented by the Portuguese population and require more thorough inspections. Thus, relying on risk assessment usig machine learning can have a positive influence in economic crime prevention related to food safety as well as in public health.