The Development of a Model to Predict Marbling Score for fattening Kamphaeng Saen Beef Breed Using Data Mining

The purpose of this research was to create a model for beef marbling score in Kamphaeng Saen beef breed using data mining techniques. The data of the fattening cattle were transformed from Kamphaeng Saen Beef Co-operative Co., Ltd. during 2015-2019. For 5 years, 1,568 head, including the fattening period (month), the age of the cattle from the shedding and wear of the teeth The weight is alive when slaughter, the carcass weight into a fresh, carcass weight, cool carcass percentage and the marbling score in the meat. To create a forecast model by using data classification techniques with Decision Tree method with C4.5 algorithm, then test the predictive model with percentage split method by dividing the data into training set 30% and test set 70%. It is found that the accuracy is 57.1949% both because there are many factors related to the formation of fat in the muscles such as varieties, age, sex and feed, especially feed is the main factor affecting the accumulation of fat in the muscles. The growth of fat that is controlled by Stearoyl-CoA desaturase (SCD) shows that the prediction of the marbling score in fattening cow muscle with data mining techniques can be used as a means of forecasting without having to wait. The cattle were finished and assessed from the remains of the cattle.