Research on Effect Evaluation of Online Advertisement Based on Resampling Method

Table 2

The Pearson correlation coefficient of training data features on September 1.

Feature	Pearson correlation coefficient
ROI on the day	0.197
Amount of total payments	0.190
Amount of total orders	0.189
Average residence time of user	0.141
Type of media platform	0.135
Ad budget	0.127
Ad type	0.127
Number of total purchase volume	0.123
Number of total search volume	0.121
Ad position	0.119
Number of total collection volume	0.112
User amount	0.106
Number of total recommendation volume	0.098
Average views of home page	0.080
Number of total register volume	0.079
User gender	0.078
Views of home page≥8	0.068
Average views of business page	0.063
Max of user age	0.061
Min of user age	0.060
Views of business page ≥ 8	0.043
Number of order people	0.040
Number of orders	0.035
Number of payers	0.022
Number of payments	0.013
Residence time of user ≥ 8 min	0.001
Views of home page = 1	−0.001
Residence time of user ≥ 1 min	−0.002
Views of business page = 1	−0.022
At the 0.01 level (double tails), the correlation is significant.
At the 0.05 level (double tails), the correlation is significant.