Prediction of Dairy Product Quality Risk Based on Extreme Learning Machine
Ruifang Zhang, Lin Zhou, Min Zuo, Qingchuan Zhang, Mingwen Bi, Qingyu Jin, Zelong Xu · 2018
The use of machine learning can study the objective laws of the data to find out the relationship between the data, but also can achieve data classification and prediction. This article is based on national food safety sampling platform for dairy products testing data for the study to predict the risk safety of dairy products. Firstly, the sampling data of dairy products was selected from the database and the data was preprocessed. Then, the characteristic data were selected and classified into these data. Finally, the key food safety risk early-warning model was constructed by using Extreme learning machine (ELM) and the kernel-based extreme learning machine (K-ELM) respectively. Commonly used BP neural network and Support Vector Machines(SVM) network contrast. After ELM and K-ELM again to compare the experimental results. From the experimental results, it can be seen that the neural network algorithm based on kernel extreme learning machine has a high accuracy in food safety prediction and the training time is short. Through the extreme learning machine effective prediction of food safety, to regulate the quality and safety of food, so as to protect the quality of food safety in China.