Method for quickly identification of cow milk and goat milk

• Main Authors: HAITONG WANG, CHENDONG YANG, SHI LIU, JIANMING YI, PINGMIN WAN, LEI CHENG, JUNQING NI, XIUZHONG HU, WENJUN LIU, XIAOHUA LIU, JUNDONG SU, AHMED SABEK, CHUNFANG LI, YABIN MA, SHUJUN ZHANG, LIANGKANG NAN, YI ZHANG, WENJUAN DU
• Format: Patent
• Language: English
• Published: 08.09.2022
• Subjects: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIRCHEMICAL OR PHYSICAL PROPERTIES; MEASURING; PHYSICS; TESTING

The present disclosure relates to a model and a method for spectral identification of cow milk and goat milk, which is related to mid-infrared spectroscopy analysis. The invention comprises the following steps: (1) selecting fresh cow :milk and. fresh. goat milk samples; (2) scanning' the samples in the mid-red spectrum range to obtain mid-infrared spectrum data; (3) preprocessing the original mid-infrared spectrum to remove outliers; (4) dividing the preprocessed data set into a training set and a test set according to the 10 principle of stratified sampling; (5) reducing the dimension of the training set to improve the speed of model training; (6) using K Nearest Neighbor (KNN) algorithm. and other algorithms on the training to build models through lO-fold cross-validation that are able to distinguish between cow 15 milk and goat milk, and using accuracy, specificity, sensitivity and AUC to evaluate and filter the models; (7) predicting the generalization ability of the optimal model.