Machine learning algorithm for estimating and optimizing the phytochemical content and physicochemical properties of okra slices in an infrared heating system

Abstract

This study investigates how different air temperatures and infrared intensities affect the physicochemical properties of dried okra at different airflow rates. The model was developed using machine learning, and Okra's physicochemical properties were optimized using a self-organizing map (SOM). The results showed that higher infrared intensity and air temperature improved rehydration and colour but reduced water activity and vitamin C levels. In contrast, faster airflow helped preserve quality by counteracting the negative effects of higher temperatures and infrared intensity. The SOM algorithm identified five optimal drying conditions, revealing that lower temperatures, lower infrared intensity, and higher airflow provided optimal conditions for improving the quality of okra slices. Interestingly, the machine learning model's predictions closely matched the test data sets, providing valuable insights for understanding and controlling the factors affecting okra drying performances. This study used machine learning to optimize the drying process of okra, a new approach for improving food drying techniques. It offers valuable insights for the food industry in its quest to improve efficiency without sacrificing product quality.