using machine learning to determine aflatoxin level in rice based on ambient temperature and relative humidity

In the quality control of imported cereals, one of the critical parameters is the level of rice contamination with aflatoxins, which is measured by high-performance liquid chromatography with an immunoaffinity column. in the present study, it is investigated the effect of temperature and relative humidity of the rice storage location on its aflatoxin level. for this purpose, aflatoxin analysis results were collected period of one year. the mean temperature and relative humidity of the storage location were recorded. the effect of ambient temperature and relative humidity on the level of aflatoxin b1 and b2 in rice was determined through mathematical modeling using a supervised machine learning method based on gradient descent. based on mathematical equations fitted to the results of aflatoxin measurements in 444 rice samples, a temperature of 25 to 35 °c and a relative humidity of 50 to 60 % are recommended to reduce aflatoxin when storing imported rice in warehouses.