Research on Classification Model of Fermented Milk Quality Control Based on Data Mining

Abstract

Fermented milk has already entered the household as a kind of health drink. With the expansion of the fermented milk market, greater demands are being placed on food producers. Therefore, improving the quality of fermented milk and reducing the customer complaint rate have become the focus of food producers.Artificial sensory evaluation will be affected by your own physical condition, and the qualitative change period sample is not suitable for artificial evaluation. Therefore, the use of electronic instruments for measurement is more efficient than traditional methods, and it is easier to maintain storage, which is conducive to analysis and allows researchers to intuitively judge quality.By establishing random forest model, LR model and AdaBoosting model, we compare the accuracy of these models to find the most suitable classification model. The results show that the method has the ability to recognize the color, aroma, taste and quality of fermented milk.The rate of confirmation is 96.8%. The experimental results show that the expected results are achieved.