Research on Quality Prediction of Ore Processing Based on Mathematical Modeling Methods
A prediction problem on the quality indicators of ore processing is studied by mathematical modeling methods.Two data tables of Excel attachment are used to divide the data into three types.Five models(linear re-gression,random forest,decision tree,XGBoost,GBDT model)are used to predict the relevant data from dif-ferent dimensions.In order to obtain the fitting accuracy of the model,the method of decision coefficient is ap-plied to quantitatively analyze the prediction effect of the five models.Naturally,the decision tree and GBDT models are selected to predict the quality indicators of ore processing and the qualification rate of ore processing,respectively.This study verifies the effectiveness and feasibility of the proposed models via experiments,which has certain practical application value for improving industrial production efficiency and product quality.
ore processingquality and qualification ratedecision treeGBDT algorithm
万方数据
维普
