Fig 1.
Functions and classification of computer data mining technology.
Fig 2.
Schematic diagram of data information processing process.
Fig 3.
Data mining system model.
Fig 4.
Marketing model based on data mining.
Fig 5.
Attempt framework of the model.
Fig 6.
Training process of model construction.
Fig 7.
Construction process of XGBoost model.
Table 1.
Selection of data set parameters.
Fig 8.
Training implementation process of the model.
Fig 9.
Evaluation architecture of XGBoost algorithm model (RF: Random forest algorithm; LR: Logical regression algorithm; DT: Decision tree algorithm).
Fig 10.
XGBoost algorithm training process.
Table 2.
Customer information database.
Fig 11.
Data preparation and indicator design process of XGBoost model.
Table 3.
Design of macro market indicators.
Fig 12.
Comparison of the performance of four algorithms.
Table 4.
Prediction results of XGBoost algorithm.
Table 5.
Comparison of enterprise sales forecast models.
Table 6.
Significance test of the difference between the results of important characteristics.
Fig 13.
Relationship between the nature of different enterprises and customer consumption.
Fig 14.
Distribution diagram of customers’ application in different channels of the enterprise.